Income and expenditure declaration data processing method and system based on multi-heterogeneous models

Abstract

The present invention discloses a method and system for processing income and expenditure declaration data based on a multi-heterogeneous model, the method comprising: obtaining the international balance of payments declaration data and the corresponding historical verification operation records sent by the target declaration end; predicting the declaration data risk corresponding to the target declaration end according to the preset multi-heterogeneous neural network model and the international balance of payments declaration data; determining the abnormal operation risk corresponding to the target declaration end according to the verification operation records of multiple similar declaration ends and the historical verification operation records; calculating the product of the declaration data risk and the abnormal operation risk to obtain the verification priori risk value corresponding to the target declaration end; the verification priori risk value is used to trigger a warning mark for the target declaration end when it is greater than a preset risk threshold. It can be seen that the present invention can analyze and warn the risks of the declaration end in advance, improve the efficiency and effect of the verification of the international balance of payments declaration data, and reduce verification errors.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for processing income and expenditure declaration data based on multiple heterogeneous models. Background Art

[0002] At present, there are many problems in the declaration and verification of international balance of payments, such as inadequate inspection of declaration data due to the experience of the declarants, differences in regional supervision, and large and cumbersome daily data volume. Traditional declarations mainly rely on manual processing operations, resulting in certain inconsistencies in inspection standards and low inspection efficiency in declaration and verification. Although some existing improvement plans have proposed the use of computer automation to verify the declaration data of international balance of payments, they have not fully utilized neural network technology to realize the early risk analysis of the declaration data, so as to warn the declaration end that may have risks in advance. Therefore, the existing declaration data processing plan has low risk warning capabilities, low efficiency, and high labor costs. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for processing income and expenditure declaration data based on multi-heterogeneous models, which can analyze and warn the risks at the declaration end in advance, improve the efficiency and effect of international balance of payments declaration data verification, and reduce verification errors.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a method for processing income and expenditure declaration data based on a multi-heterogeneous model, the method comprising:

[0005] Obtain the balance of payments declaration data and corresponding historical verification operation records sent by the target declaration terminal;

[0006] According to the preset multi-heterogeneous neural network model and the balance of payments declaration data, predict the declaration data risk corresponding to the target declaration end;

[0007] Determine the abnormal operation risk corresponding to the target reporting terminal according to the verification operation records of multiple similar reporting terminals and the historical verification operation records;

[0008] The product of the declared data risk and the abnormal operation risk is calculated to obtain a verification a priori risk value corresponding to the target declaration end; the verification a priori risk value is used to trigger a warning mark for the target declaration end when it is greater than a preset risk threshold.

[0009] As an optional embodiment, in the first aspect of the present invention, the historical verification operation records include manual verification operation records of the data of the target reporting end being manually verified in multiple historical time periods; the manual verification operation records are manual interaction operation records on the web page or software end recorded by screen recording software.

[0010] As an optional embodiment, in the first aspect of the present invention, the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network and an operation rationality prediction network.

[0011] As an optional embodiment, in the first aspect of the present invention, the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained using a training data set including multiple training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict specific operation execution commands based on the automated operation process generated by the automated operation prediction network and send them to the corresponding computer device for execution, and the automated operation execution network is trained using a training data set including multiple training automated operation processes and corresponding operation execution command annotations.

[0012] As an optional embodiment, in the first aspect of the present invention, the verification result risk prediction network is used to obtain the execution result of the computer device executing the operation execution command and predict the data risk probability corresponding to the execution result, and the verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters, and the operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations.

[0013] As an optional implementation, in the first aspect of the present invention, predicting the risk of the declaration data corresponding to the target declaration terminal according to the preset multi-heterogeneous neural network model and the balance of payments declaration data includes:

[0014] Inputting the balance of payments declaration data into the multi-heterogeneous neural network model to obtain corresponding data risk probability and rationality parameters;

[0015] calculating a probability weight proportional to the plausibility parameter;

[0016] The product of the data risk probability and the probability weight is calculated to obtain the declared data risk corresponding to the target declaration end.

[0017] As an optional implementation, in the first aspect of the present invention, determining the abnormal operation risk corresponding to the target reporting terminal according to the verification operation records of multiple similar reporting terminals and the historical verification operation records includes:

[0018] Inputting the historical verification operation record into a preset image recognition neural network to obtain a specific operation information sequence corresponding to the historical verification operation record; the specific operation information sequence includes operation types, operation positions and durations of multiple verification operations;

[0019] According to the device information of the target reporting terminal, based on the similarity algorithm, multiple similar reporting terminals are determined from multiple candidate reporting terminals; the information similarity between the device information of the similar reporting terminals and the device information of the target reporting terminal is greater than a preset similarity threshold;

[0020] Calculating the operation similarity between the verification operation record of each similar declaring terminal and the specific operation information sequence;

[0021] According to the operation similarities of all the similar reporting terminals, the abnormal operation risk corresponding to the target reporting terminal is calculated.

[0022] As an optional implementation, in the first aspect of the present invention, the calculating the abnormal operation risk corresponding to the target reporting terminal according to the operation similarities of all the similar reporting terminals includes:

[0023] Calculate the weighted sum average of the operation similarities of all the similar reporting terminals to obtain a similarity parameter; wherein the weighted calculation weight corresponding to each operation similarity is proportional to the information similarity corresponding to the corresponding similar reporting terminal;

[0024] Calculate the ratio of 1 to the similarity parameter to obtain the abnormal operation risk corresponding to the target reporting end.

[0025] A second aspect of an embodiment of the present invention discloses a revenue and expenditure declaration data processing system based on a multi-heterogeneous model, the system comprising:

[0026] The acquisition module is used to obtain the balance of payments declaration data and corresponding historical verification operation records sent by the target declaration terminal;

[0027] A prediction module, used for predicting the declaration data risk corresponding to the target declaration terminal according to a preset multi-heterogeneous neural network model and the balance of payments declaration data;

[0028] A determination module, used to determine the abnormal operation risk corresponding to the target reporting terminal according to the verification operation records of multiple similar reporting terminals and the historical verification operation records;

[0029] A calculation module is used to calculate the product of the declared data risk and the abnormal operation risk to obtain a verification priori risk value corresponding to the target declaration end; the verification priori risk value is used to trigger a warning mark for the target declaration end when it is greater than a preset risk threshold.

[0030] As an optional embodiment, in the second aspect of the present invention, the historical verification operation records include manual verification operation records of the data of the target reporting end that have been manually verified in multiple historical time periods; the manual verification operation records are manual interaction operation records on the web page or software end recorded by screen recording software.

[0031] As an optional embodiment, in the second aspect of the present invention, the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network and an operation rationality prediction network.

[0032] As an optional embodiment, in the second aspect of the present invention, the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained using a training data set including multiple training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict specific operation execution commands based on the automated operation process generated by the automated operation prediction network and send them to the corresponding computer device for execution, and the automated operation execution network is trained using a training data set including multiple training automated operation processes and corresponding operation execution command annotations.

[0033] As an optional embodiment, in the second aspect of the present invention, the verification result risk prediction network is used to obtain the execution result of the computer device executing the operation execution command and predict the data risk probability corresponding to the execution result, and the verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters, and the operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations.

[0034] As an optional implementation, in the second aspect of the present invention, the prediction module predicts the specific manner of the declaration data risk corresponding to the target declaration end according to the preset multi-heterogeneous neural network model and the balance of payments declaration data, including:

[0035] Inputting the balance of payments declaration data into the multi-heterogeneous neural network model to obtain corresponding data risk probability and rationality parameters;

[0036] calculating a probability weight proportional to the plausibility parameter;

[0037] The product of the data risk probability and the probability weight is calculated to obtain the declared data risk corresponding to the target declaration end.

[0038] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module determines the abnormal operation risk corresponding to the target declaration terminal according to the verification operation records of multiple similar declaration terminals and the historical verification operation records includes:

[0039] Inputting the historical verification operation record into a preset image recognition neural network to obtain a specific operation information sequence corresponding to the historical verification operation record; the specific operation information sequence includes operation types, operation positions and durations of multiple verification operations;

[0040] According to the device information of the target reporting terminal, based on the similarity algorithm, multiple similar reporting terminals are determined from multiple candidate reporting terminals; the information similarity between the device information of the similar reporting terminals and the device information of the target reporting terminal is greater than a preset similarity threshold;

[0041] Calculating the operation similarity between the verification operation record of each similar declaring terminal and the specific operation information sequence;

[0042] According to the operation similarities of all the similar reporting terminals, the abnormal operation risk corresponding to the target reporting terminal is calculated.

[0043] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module calculates the abnormal operation risk corresponding to the target reporting terminal according to the operation similarities of all the similar reporting terminals includes:

[0044] Calculate the weighted sum average of the operation similarities of all the similar reporting terminals to obtain a similarity parameter; wherein the weighted calculation weight corresponding to each operation similarity is proportional to the information similarity corresponding to the corresponding similar reporting terminal;

[0045] Calculate the ratio of 1 to the similarity parameter to obtain the abnormal operation risk corresponding to the target reporting end.

[0046] The third aspect of the present invention discloses another revenue and expenditure declaration data processing system based on a multi-heterogeneous model, the system comprising:

[0047] A memory storing executable program code;

[0048] a processor coupled to the memory;

[0049] The processor calls the executable program code stored in the memory to execute part or all of the steps in the method for processing income and expenditure declaration data based on multi-heterogeneous models disclosed in the first aspect of the present invention.

[0050] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute part or all of the steps in the income and expenditure declaration data processing method based on multi-heterogeneous models disclosed in the first aspect of the present invention.

[0051] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0052] The present invention can predict the risk of declaration data corresponding to the declaration end based on a preset multi-heterogeneous neural network model and international balance of payments declaration data, and then determine the abnormal operation risk corresponding to the declaration end based on the verification operation records of multiple similar declaration ends, so as to comprehensively calculate the prior risk value of the declaration end and make warning marks in time, so as to analyze and warn the risks of the declaration end in advance, improve the efficiency and effect of the verification of international balance of payments declaration data, and reduce verification errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 It is a flow chart of a method for processing income and expenditure declaration data based on multiple heterogeneous models disclosed in an embodiment of the present invention.

[0055] Figure 2 It is a structural schematic diagram of a revenue and expenditure declaration data processing system based on multiple heterogeneous models disclosed in an embodiment of the present invention.

[0056] Figure 3 It is a structural schematic diagram of another income and expenditure declaration data processing system based on multiple heterogeneous models disclosed in an embodiment of the present invention.

[0057] Figure 4 The present invention discloses an operating interface of a revenue and expenditure declaration data processing software.

[0058] Figure 5It is a schematic diagram of an operation interface of another income and expenditure declaration data processing software disclosed in an embodiment of the present invention.

[0059] Figure 6 It is a schematic diagram of an operation interface of another income and expenditure declaration data processing software disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0062] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0063] The present invention discloses a method and system for processing income and expenditure declaration data based on a multi-heterogeneous model, which can predict the declaration data risk corresponding to the declaration end based on a preset multi-heterogeneous neural network model and international balance of payments declaration data, and then determine the abnormal operation risk corresponding to the declaration end based on the verification operation records of multiple similar declaration ends, so as to comprehensively calculate the prior risk value of the declaration end and make warning marks in time, so as to analyze and warn the risks of the declaration end in advance, improve the efficiency and effect of the verification of international balance of payments declaration data, and reduce verification errors. The following are detailed descriptions.

[0064] Embodiment 1

[0065] See also Figure 1 , Figure 1This is a flow chart of a method for processing income and expenditure declaration data based on a multi-heterogeneous model disclosed in an embodiment of the present invention. Figure 1 The described method for processing revenue and expenditure declaration data based on multiple heterogeneous models can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the revenue and expenditure declaration data processing method based on multiple heterogeneous models may include the following operations:

[0066] 101. Obtain the balance of payments declaration data and corresponding historical verification operation records sent by the target declaration terminal.

[0067] 102. Based on the preset multi-heterogeneous neural network model and the balance of payments declaration data, predict the declaration data risk corresponding to the target declaration end.

[0068] 103. Based on the verification operation records of multiple similar reporting terminals and historical verification operation records, determine the abnormal operation risk corresponding to the target reporting terminal.

[0069] 104. Calculate the product of the declared data risk and the abnormal operation risk to obtain the verification prior risk value corresponding to the target declaration end.

[0070] Optionally, the verification prior risk value is used to trigger a warning mark on the target reporting end when it is greater than a preset risk threshold.

[0071] It can be seen that the above-mentioned embodiments of the invention can predict the declaration data risk corresponding to the declaration end based on the preset multi-heterogeneous neural network model and the balance of payments declaration data, and then determine the abnormal operation risk corresponding to the declaration end based on the verification operation records of multiple similar declaration ends, so as to comprehensively calculate the prior risk value of the declaration end and make warning marks in time, so as to analyze and warn the risks of the declaration end in advance, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0072] As an optional embodiment, in the above steps, the historical verification operation records include manual verification operation records of the data of the target reporting end being manually verified in multiple historical time periods; the manual verification operation records are manual interaction operation records on the web page or software end recorded by screen recording software.

[0073] It can be seen that through the above optional embodiments, the content of the historical verification operation records is clarified, which can more accurately represent the manual verification records of the reporting end in the historical time period, so as to accurately analyze the abnormal operation risks, assist in the early analysis and warning of the risks of the reporting end, improve the efficiency and effectiveness of the verification of international balance of payments declaration data, and reduce verification errors.

[0074] As an optional embodiment, in the above steps, the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network and an operation rationality prediction network.

[0075] It can be seen that through the above optional embodiments, the model network composition details of the multi-heterogeneous neural network model are clarified, which can more accurately predict the risks of the declared data, accurately evaluate the prior risks in the subsequent period, and assist in analyzing and warning the risks at the declaration end in advance, thereby improving the efficiency and effectiveness of the verification of the balance of payments declaration data and reducing verification errors.

[0076] As an optional embodiment, in the above steps, the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained by a training data set including multiple training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict specific operation execution commands according to the automated operation process generated by the automated operation prediction network and send them to the corresponding computer device for execution, and the automated operation execution network is trained by a training data set including multiple training automated operation processes and corresponding operation execution command annotations.

[0077] It can be seen that through the above-mentioned optional embodiments, the network details of the automated operation prediction network and the automated operation execution network are clarified, which can achieve more accurate prediction of the automated verification operation of the declared data, so as to accurately assess the verification risk of the data in the subsequent period, assist in analyzing and warning the risks at the declaring end in advance, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0078] As an optional embodiment, in the above steps, the verification result risk prediction network is used to obtain the execution result of the computer device executing the operation execution command and predict the data risk probability corresponding to the execution result, and the verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters, and the operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations.

[0079] It can be seen that through the above-mentioned optional embodiments, the network details of the verification result risk prediction network and the operation rationality prediction network are clarified, which can achieve a more accurate prediction of the risk and rationality of the automated verification results of the declared data, so as to accurately evaluate the verification risk of the data in the subsequent period, assist in the early analysis and warning of the risks on the reporting end, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0080] As an optional embodiment, in the above steps, predicting the declaration data risk corresponding to the target declaration end according to the preset multi-heterogeneous neural network model and the balance of payments declaration data includes:

[0081] Input the balance of payments declaration data into the multi-heterogeneous neural network model to obtain the corresponding data risk probability and rationality parameters;

[0082] Calculate probability weights proportional to plausibility parameters;

[0083] Calculate the product of the data risk probability and the probability weight to obtain the reporting data risk corresponding to the target reporting end.

[0084] It can be seen that through the above-mentioned optional embodiments, the data risk probability and rationality parameters corresponding to the declaration data submitted by the declaring end can be predicted through a multi-heterogeneous neural network model, and the declaration data risk corresponding to the declaring end can be accurately predicted through weight product calculation, which is convenient for subsequent accurate analysis of the prior risk of the declaring end, and assists in realizing early analysis and warning of the risks of the declaring end, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0085] As an optional embodiment, in the above steps, determining the abnormal operation risk corresponding to the target reporting end according to the verification operation records of multiple similar reporting ends and the historical verification operation records includes:

[0086] Inputting the historical verification operation records into a preset image recognition neural network to obtain a specific operation information sequence corresponding to the historical verification operation records; optionally, the specific operation information sequence includes operation types, operation positions and durations of multiple verification operations;

[0087] According to the device information of the target reporting terminal, based on the similarity algorithm, multiple similar reporting terminals are determined from multiple candidate reporting terminals; optionally, the information similarity between the device information of the similar reporting terminal and the device information of the target reporting terminal is greater than a preset similarity threshold;

[0088] Calculate the operational similarity between the verification operation record of each similar declaration terminal and the specific operation information sequence;

[0089] Based on the operation similarities of all similar reporting terminals, calculate the abnormal operation risk corresponding to the target reporting terminal.

[0090] It can be seen that through the above optional embodiments, the specific operation information sequence corresponding to the operation record screen recording can be determined through the image recognition neural network, and similar reporting terminals are first screened out based on the similarity calculation, and then the abnormal operation risk corresponding to the target reporting terminal is determined through the similarity calculation of the verification operation records of the similar reporting terminals and the specific operation information sequence of the target reporting terminal, which is convenient for the subsequent accurate analysis of the prior risks of the reporting terminal, and assists in realizing the early analysis and warning of the risks of the reporting terminal, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0091] As an optional embodiment, in the above steps, calculating the abnormal operation risk corresponding to the target reporting end according to the operation similarity of all similar reporting ends includes:

[0092] Calculate the weighted sum average of the operation similarities of all similar reporting terminals to obtain a similarity parameter; optionally, the weighted calculation weight corresponding to each operation similarity is proportional to the information similarity corresponding to the corresponding similar reporting terminal;

[0093] Calculate the ratio of 1 to the similarity parameter to obtain the abnormal operation risk corresponding to the target reporting end.

[0094] It can be seen that through the above-mentioned optional embodiments, the weighted sum and average calculation and the inverse calculation can be performed on the operation similarities of all similar reporting terminals to accurately determine the abnormal operation risk corresponding to the target reporting terminal, facilitate the subsequent accurate analysis of the prior risk of the reporting terminal, and assist in realizing the early analysis and warning of the risks of the reporting terminal, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0095] In a specific implementation scheme, the above disclosed technical solution is implemented as an international balance of payments declaration data processing software, and its login interface is as follows: Figure 4 As shown, the administrator account can log in to process the data submitted by each reporting end, and the initial management interface for the data submitted by the reporting end is as follows Figure 5 As shown, examples of income and expenditure declaration data submitted by each declaration terminal can be referred to Figure 6 The back-end server of the software is deployed with the multi-heterogeneous neural network algorithm disclosed in the present invention to assist the administrator in identifying and labeling the risks of the reporting end, so as to facilitate the administrator's knowledge and improve work efficiency.

[0096] Embodiment 2

[0097] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of a revenue and expenditure declaration data processing system based on a multi-heterogeneous model disclosed in an embodiment of the present invention. Figure 2The described revenue and expenditure declaration data processing system based on multiple heterogeneous models can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the revenue and expenditure declaration data processing system based on multiple heterogeneous models may include:

[0098] The acquisition module 201 is used to obtain the international balance of payments declaration data and the corresponding historical verification operation records sent by the target declaration terminal.

[0099] The prediction module 202 is used to predict the declaration data risk corresponding to the target declaration end based on the preset multi-heterogeneous neural network model and the balance of payments declaration data.

[0100] The determination module 203 is used to determine the abnormal operation risk corresponding to the target reporting end based on the verification operation records of multiple similar reporting ends and the historical verification operation records.

[0101] The calculation module 204 is used to calculate the product of the declared data risk and the abnormal operation risk to obtain the verification prior risk value corresponding to the target declaration end.

[0102] Optionally, the verification prior risk value is used to trigger a warning mark on the target reporting end when it is greater than a preset risk threshold.

[0103] It can be seen that the above-mentioned embodiments of the invention can predict the declaration data risk corresponding to the declaration end based on the preset multi-heterogeneous neural network model and the balance of payments declaration data, and then determine the abnormal operation risk corresponding to the declaration end based on the verification operation records of multiple similar declaration ends, so as to comprehensively calculate the prior risk value of the declaration end and make warning marks in time, so as to analyze and warn the risks of the declaration end in advance, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0104] As an optional embodiment, the historical verification operation records include manual verification operation records of the data of the target reporting end that were manually verified in multiple historical time periods; the manual verification operation records are manual interactive operation records on the web page or software end recorded by screen recording software.

[0105] It can be seen that through the above optional embodiments, the content of the historical verification operation records is clarified, which can more accurately represent the manual verification records of the reporting end in the historical time period, so as to accurately analyze the abnormal operation risks, assist in the early analysis and warning of the risks of the reporting end, improve the efficiency and effectiveness of the verification of international balance of payments declaration data, and reduce verification errors.

[0106] As an optional embodiment, the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network, and an operation rationality prediction network.

[0107] It can be seen that through the above optional embodiments, the model network composition details of the multi-heterogeneous neural network model are clarified, which can more accurately predict the risks of the declared data, accurately evaluate the prior risks in the subsequent period, and assist in analyzing and warning the risks at the declaration end in advance, thereby improving the efficiency and effectiveness of the verification of the balance of payments declaration data and reducing verification errors.

[0108] As an optional embodiment, the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained using a training data set including multiple training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict specific operation execution commands based on the automated operation process generated by the automated operation prediction network and send them to the corresponding computer device for execution, and the automated operation execution network is trained using a training data set including multiple training automated operation processes and corresponding operation execution command annotations.

[0109] It can be seen that through the above-mentioned optional embodiments, the network details of the automated operation prediction network and the automated operation execution network are clarified, which can achieve more accurate prediction of the automated verification operation of the declared data, so as to accurately assess the verification risk of the data in the subsequent period, assist in analyzing and warning the risks at the declaring end in advance, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0110] As an optional embodiment, the verification result risk prediction network is used to obtain the execution results of the operation execution command executed by the computer device and predict the data risk probability corresponding to the execution result. The verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters. The operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations.

[0111] It can be seen that through the above-mentioned optional embodiments, the network details of the verification result risk prediction network and the operation rationality prediction network are clarified, which can achieve a more accurate prediction of the risk and rationality of the automated verification results of the declared data, so as to accurately evaluate the verification risk of the data in the subsequent period, assist in the early analysis and warning of the risks on the reporting end, improve the efficiency and effectiveness of the verification of the balance of payments declaration data, and reduce verification errors.

[0112] As an optional embodiment, the prediction module predicts the specific method of the declaration data risk corresponding to the target declaration end according to the preset multi-heterogeneous neural network model and the balance of payments declaration data, including:

[0113] Input the balance of payments declaration data into the multi-heterogeneous neural network model to obtain the corresponding data risk probability and rationality parameters;

[0114] Calculate probability weights proportional to plausibility parameters;

[0115] Calculate the product of the data risk probability and the probability weight to obtain the reporting data risk corresponding to the target reporting end.

[0116] It can be seen that through the above-mentioned optional embodiments, the data risk probability and rationality parameters corresponding to the declaration data submitted by the declaring end can be predicted through a multi-heterogeneous neural network model, and the declaration data risk corresponding to the declaring end can be accurately predicted through weight product calculation, which is convenient for subsequent accurate analysis of the prior risk of the declaring end, and assists in realizing early analysis and warning of the risks of the declaring end, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0117] As an optional embodiment, the specific manner in which the determination module determines the abnormal operation risk corresponding to the target declaration terminal according to the verification operation records of multiple similar declaration terminals and the historical verification operation records includes:

[0118] Inputting the historical verification operation records into a preset image recognition neural network to obtain a specific operation information sequence corresponding to the historical verification operation records; optionally, the specific operation information sequence includes operation types, operation positions and durations of multiple verification operations;

[0119] According to the device information of the target reporting terminal, based on the similarity algorithm, multiple similar reporting terminals are determined from multiple candidate reporting terminals; optionally, the information similarity between the device information of the similar reporting terminal and the device information of the target reporting terminal is greater than a preset similarity threshold;

[0120] Calculate the operational similarity between the verification operation record of each similar declaration terminal and the specific operation information sequence;

[0121] Based on the operation similarities of all similar reporting terminals, calculate the abnormal operation risk corresponding to the target reporting terminal.

[0122] It can be seen that through the above optional embodiments, the specific operation information sequence corresponding to the operation record screen recording can be determined through the image recognition neural network, and similar reporting terminals are first screened out based on the similarity calculation, and then the abnormal operation risk corresponding to the target reporting terminal is determined through the similarity calculation of the verification operation records of the similar reporting terminals and the specific operation information sequence of the target reporting terminal, which is convenient for the subsequent accurate analysis of the prior risks of the reporting terminal, and assists in realizing the early analysis and warning of the risks of the reporting terminal, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0123] As an optional embodiment, the specific method in which the determination module calculates the abnormal operation risk corresponding to the target reporting terminal according to the operation similarity of all similar reporting terminals includes:

[0124] Calculate the weighted sum average of the operation similarities of all similar reporting terminals to obtain a similarity parameter; optionally, the weighted calculation weight corresponding to each operation similarity is proportional to the information similarity corresponding to the corresponding similar reporting terminal;

[0125] Calculate the ratio of 1 to the similarity parameter to obtain the abnormal operation risk corresponding to the target reporting end.

[0126] It can be seen that through the above-mentioned optional embodiments, the weighted sum and average calculation and the inverse calculation can be performed on the operation similarities of all similar reporting terminals to accurately determine the abnormal operation risk corresponding to the target reporting terminal, facilitate the subsequent accurate analysis of the prior risk of the reporting terminal, and assist in realizing the early analysis and warning of the risks of the reporting terminal, thereby improving the efficiency and effectiveness of the verification of international balance of payments declaration data and reducing verification errors.

[0127] Embodiment 3

[0128] See also Figure 3 , Figure 3 It is another income and expenditure declaration data processing system based on multi-heterogeneous models disclosed in an embodiment of the present invention. Figure 3 The described revenue and expenditure declaration data processing system based on multiple heterogeneous models is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the revenue and expenditure declaration data processing system based on multiple heterogeneous models may include:

[0129] A memory 301 storing executable program codes;

[0130] a processor 302 coupled to the memory 301;

[0131] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the method for processing income and expenditure declaration data based on multi-heterogeneous models described in the first embodiment.

[0132] Embodiment 4

[0133] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the method for processing revenue and expenditure declaration data based on multi-heterogeneous models described in the first embodiment.

[0134] Embodiment 5

[0135] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the method for processing income and expenditure declaration data based on multi-heterogeneous models described in Example 1.

[0136] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0137] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0138] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0139] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0140] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0141] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0142] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0143] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0144] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0145] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0146] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0147] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0148] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0149] Finally, it should be noted that the method and system for processing income and expenditure declaration data based on multi-heterogeneous models disclosed in the embodiment of the present invention only disclose the preferred embodiments of the present invention, which are only used to illustrate the technical scheme of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical schemes described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical schemes from the spirit and scope of the technical schemes of the embodiments of the present invention.

Claims

1. A method for processing income and expenditure declaration data based on multi-heterogeneous models, characterized in that: The method comprises: Obtain the balance of payments declaration data and corresponding historical verification operation records sent by the target declaration terminal; According to the preset multi-heterogeneous neural network model and the balance of payments declaration data, the declaration data risk corresponding to the target declaration end is predicted, including: Inputting the balance of payments declaration data into the multi-heterogeneous neural network model to obtain corresponding data risk probability and rationality parameters; calculating a probability weight proportional to the plausibility parameter; The product of the data risk probability and the probability weight is calculated to obtain the declaration data risk corresponding to the target declaration end; the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network and an operation rationality prediction network; the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained by a training data set including a plurality of training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict a specific operation execution command according to the automated operation process generated by the automated operation prediction network and send it to the corresponding The computer device is used to execute the operation, and the automated operation execution network is trained by a training data set including multiple training automated operation processes and corresponding operation execution command annotations; the verification result risk prediction network is used to obtain the execution result of the operation execution command executed by the computer device and predict the data risk probability corresponding to the execution result, and the verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters, and the operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations; Determine the abnormal operation risk corresponding to the target reporting terminal according to the verification operation records of multiple similar reporting terminals and the historical verification operation records; The product of the declared data risk and the abnormal operation risk is calculated to obtain a verification a priori risk value corresponding to the target declaration end; the verification a priori risk value is used to trigger a warning mark for the target declaration end when it is greater than a preset risk threshold.

2. The method for processing revenue and expenditure declaration data based on multi-heterogeneous models according to claim 1 is characterized in that: The historical verification operation records include manual verification operation records of the data of the target reporting end that were manually verified in multiple historical time periods; the manual verification operation records are manual interactive operation records on the web page or software end recorded by screen recording software.

3. The method for processing revenue and expenditure declaration data based on multi-heterogeneous models according to claim 2 is characterized in that: The determining, based on the verification operation records of a plurality of similar declaration terminals and the historical verification operation records, the abnormal operation risk corresponding to the target declaration terminal includes: Inputting the historical verification operation record into a preset image recognition neural network to obtain a specific operation information sequence corresponding to the historical verification operation record; the specific operation information sequence includes operation types, operation positions and durations of multiple verification operations; According to the device information of the target reporting terminal, based on the similarity algorithm, multiple similar reporting terminals are determined from multiple candidate reporting terminals; the information similarity between the device information of the similar reporting terminals and the device information of the target reporting terminal is greater than a preset similarity threshold; Calculating the operation similarity between the verification operation record of each similar declaring terminal and the specific operation information sequence; According to the operation similarities of all the similar reporting terminals, the abnormal operation risk corresponding to the target reporting terminal is calculated.

4. The method for processing revenue and expenditure declaration data based on multiple heterogeneous models according to claim 3 is characterized in that: The calculating the abnormal operation risk corresponding to the target reporting terminal according to the operation similarity of all the similar reporting terminals includes: Calculate the weighted sum average of the operation similarities of all the similar reporting terminals to obtain a similarity parameter; wherein the weighted calculation weight corresponding to each operation similarity is proportional to the information similarity corresponding to the corresponding similar reporting terminal; Calculate the ratio of 1 to the similarity parameter to obtain the abnormal operation risk corresponding to the target reporting end.

5. A revenue and expenditure declaration data processing system based on multiple heterogeneous models, characterized in that: The system comprises: The acquisition module is used to obtain the balance of payments declaration data and corresponding historical verification operation records sent by the target declaration terminal; A prediction module is used to predict the declaration data risk corresponding to the target declaration terminal according to a preset multi-heterogeneous neural network model and the balance of payments declaration data, including: Inputting the balance of payments declaration data into the multi-heterogeneous neural network model to obtain corresponding data risk probability and rationality parameters; calculating a probability weight proportional to the plausibility parameter; The product of the data risk probability and the probability weight is calculated to obtain the declaration data risk corresponding to the target declaration end; the multi-heterogeneous neural network model includes an interconnected automated operation prediction network, an automated operation execution network, a verification result risk prediction network and an operation rationality prediction network; the automated operation prediction network is used to predict the automated operation process corresponding to the input balance of payments declaration data, and the automated operation prediction network is trained by a training data set including a plurality of training declaration data and corresponding automated verification operation process annotations; the automated operation execution network is used to predict a specific operation execution command according to the automated operation process generated by the automated operation prediction network and send it to the corresponding The computer device is used to execute the operation, and the automated operation execution network is trained by a training data set including multiple training automated operation processes and corresponding operation execution command annotations; the verification result risk prediction network is used to obtain the execution result of the operation execution command executed by the computer device and predict the data risk probability corresponding to the execution result, and the verification result risk prediction network is trained by a training data set including multiple training verification results and corresponding risk annotations; the operation rationality prediction network is used to obtain the automated operation process and predict the corresponding rationality parameters, and the operation rationality prediction network is trained by a training data set including multiple training automated operation processes and corresponding rationality annotations; A determination module, used to determine the abnormal operation risk corresponding to the target reporting terminal according to the verification operation records of multiple similar reporting terminals and the historical verification operation records; A calculation module is used to calculate the product of the declared data risk and the abnormal operation risk to obtain a verification priori risk value corresponding to the target declaration end; the verification priori risk value is used to trigger a warning mark for the target declaration end when it is greater than a preset risk threshold.

6. A revenue and expenditure declaration data processing system based on multiple heterogeneous models, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the income and expenditure declaration data processing method based on multi-heterogeneous models as described in any one of claims 1-4.

Images