Quick withdrawal method and system for binding bank card to third-party payment platform

Abstract

The invention discloses a rapid withdrawal method and system for binding a bank card on a third-party payment platform, and the method comprises the steps: collecting the withdrawal data and bank data of the third-party payment platform, and carrying out the preprocessing of the withdrawal data and the correlation; the withdraw data comprises third-party payment platform information data and user withdraw data; performing multi-modal risk assessment on the withdrawal data to obtain a risk degree, and performing multi-modal identity verification according to bank data by adopting biological recognition to obtain first verification data; performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk degree, the first verification data and the second verification data; and constructing a third-party payment platform rapid withdrawal model according to the verification function, optimizing the third-party payment platform rapid withdrawal model according to a response error, and outputting a withdrawal result. The method not only improves the precision of rapid withdrawal, but also has good interpretability, and can be directly used in a rapid withdrawal system.

Description

Technical Field

[0001] The present invention relates to the field of quick cash withdrawal, and in particular to a quick cash withdrawal method and system for binding a bank card to a third-party payment platform. Background Art

[0002] With the rapid development of Internet technology, third-party payment platforms have become an indispensable payment tool in people's daily lives. Users use third-party payment platforms to conduct transactions, transfers, shopping and other operations, which greatly facilitates daily life. However, when users need to withdraw the balance of third-party payment platforms to bank cards, they often need to perform cumbersome operations and wait for a long time to process, which reduces the user experience.

[0003] In the traditional withdrawal process, users need to manually enter bank card information, withdrawal amount, etc., and then the third-party payment platform will check the input information one by one. During this process, the platform needs to spend a lot of time to verify information and conduct risk assessment to ensure the security of the transaction. At the same time, users also need to wait for the platform to review and approve before they can receive the withdrawal amount, which undoubtedly increases the user's waiting time.

[0004] In order to solve the above problems, existing technologies have begun to try to optimize the withdrawal process through automated means. For example, some payment platforms have introduced OCR (optical character recognition) technology to automatically identify the information in the bank card pictures uploaded by users, thereby reducing the errors and time-consuming manual input by users. However, these technologies still remain at the level of preliminary identification and processing of information, and have not fundamentally solved the problem of cumbersome and time-consuming withdrawal processes. Therefore, Kou Dai invented a new third-party payment platform to bind bank cards for quick withdrawals, which improves the security, efficiency and accuracy of withdrawals. Summary of the invention

[0005] The purpose of the present invention is to provide a fast cash withdrawal method for a third-party payment platform bound bank card.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] The present invention comprises the following steps:

[0008] Collecting withdrawal data and bank data from a third-party payment platform, and pre-processing the withdrawal data and related information; the withdrawal data includes third-party payment platform information data and user withdrawal data;

[0009] Performing a multimodal risk assessment on the withdrawal data to obtain a risk level, and using biometrics to perform a multimodal identity verification based on bank data to obtain first verification data;

[0010] Performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data, and the second verification data;

[0011] A third-party payment platform quick withdrawal model is constructed according to the verification function, the third-party payment platform quick withdrawal model is optimized according to the response error, and the withdrawal result is output.

[0012] Furthermore, the method of performing multimodal risk assessment on the withdrawal data to obtain a risk degree includes:

[0013] The evaluation indicators are the transaction amount, transaction location, transaction time, transaction type, fund destination and transaction location image of the withdrawal data;

[0014] Based on the historical data of the evaluation indicators, a deep learning algorithm is used to extract the user's behavior habits as a reference object; the evaluation indicators are subjected to real-time feature extraction to obtain evaluation features and evaluation feature values;

[0015] Normalize the evaluation features and evaluation feature values, and calculate the difference of the evaluation features based on the reference object:

[0016]

[0017] The difference of the i-th evaluation feature is The i-th evaluation feature is b i , evaluate feature b i The reference object is b o , the error constant is σ, the optimization coefficient is ζ, and the evaluation feature b i and reference object b o The matching degree is

[0018]

[0019] Calculate the risk of the assessment feature:

[0020]

[0021] The risk of the i-th evaluation feature is K c , the scale factor is λ, and the adjustment coefficient of the i-th evaluation feature is δ i , the correlation between the i-th assessment feature and risk is The number of evaluation features is m c , evaluate feature b i The characteristic value of i |;

[0022] Calculate the risk association of the assessed features:

[0023]

[0024] The risk association of the i-th evaluation feature is θ i , difference The probability of the risk of cash withdrawal occurring at the same time is

[0025] The evaluation features with a risk correlation greater than 0.3061 are taken as key features, and the risk is calculated based on the key features:

[0026]

[0027] The risk of the cth evaluation is K c , the number of key features is m c1 .

[0028] Furthermore, the method of using biometrics to perform multimodal identity verification based on bank data to obtain first verification data includes:

[0029] Input the fingerprint and face image of the withdrawal data into a multimodal biometric recognition model based on deep learning to obtain multimodal features;

[0030] Among them, the multimodal biometric recognition algorithm: by using lightweight means such as group convolution and channel shuffling, the ghost network and shuffle network lightweight network models are structurally improved; the CBAM attention mechanism is introduced, and the improved PLAM mechanism is proposed to form a multimodal biometric recognition model;

[0031] Obtain the user's identity information database and calculate the matching degree between the user features in the identity information database and the multimodal features:

[0032]

[0033] The ath genetic coefficient is ζ a , the ath user feature is u a , the a-th multimodal feature is d a , the matching degree of the sth identity verification is E s , the authentication is correct cr, user characteristics u a The joint probability distribution of the correct authentication cr is p(u a ,cr), user feature u a The marginal distribution of a ), the marginal distribution of the correct identity authentication cr is p(cr), the decision function is γ(·), and the user feature u a and multimodal features d a The cosine similarity is The number of matching features is N1, and the value of cosine similarity is

[0034] The output of the matching degree greater than 0.658 is considered as identity verification success, otherwise it is considered as identity verification failure, where identity verification success is 1 and identity verification failure is 0;

[0035] The identity verification result is output as the first verification data.

[0036] Furthermore, the method of constructing a verification function according to the risk degree, the first verification data and the second verification data includes:

[0037]

[0038] The verification function is The first verification data is H1, the second verification data is H2, and the risk level of the cth evaluation is K c ;

[0039] When the risk level is greater than 0.682, a danger warning is triggered and the account is locked, and a password verification is performed. The password verification includes a danger warning option and a no-risk option.

[0040] When the verification function is 0, it means that withdrawals are not possible; when the verification function is not 0, if the verification risk is within a safe range, withdrawals can be made, otherwise withdrawals cannot be made.

[0041] Furthermore, a method for constructing a third-party payment platform quick cash withdrawal model according to the verification function includes:

[0042] According to the predicted withdrawal result obtained by the verification function, the minimum loss function of the actual withdrawal result and the predicted withdrawal result is used as the objective function of the third-party payment platform's fast withdrawal model;

[0043] The third-party payment platform's fast cash withdrawal model includes density clustering algorithm, autoencoder, and neural network algorithm;

[0044] The density clustering algorithm identifies the high-density and low-density areas of the input data, and regards the points in the high-density area as normal data, and the points in the low-density area or on the boundary as abnormal data;

[0045] The autoencoder encodes normal data into a low-dimensional hidden layer space through unsupervised learning, and then reconstructs it back to the original input space, learns the compressed representation of normal data, and uses the compressed representation as the extraction feature;

[0046] The neural network algorithm learns withdrawal characteristics and patterns, performs real-time classification and risk assessment based on withdrawal requests, and uses verification functions based on risk assessment to make fast withdrawal decisions on third-party payment platforms.

[0047] Furthermore, the method for optimizing the third-party payment platform quick cash withdrawal model according to the response error includes:

[0048] The swarm optimization algorithm is introduced to perform chaotic mapping on the particle population. The expression is:

[0049] k w+1 =k w +β·(2r1-1)·(1-k w )

[0050] The chaotic mapping population sequence of the w+1th particle is k a+1 , the initial population sequence of the w+1th particle is k w , the chaotic variation factor is β, and the random number from 0 to 1 is r1;

[0051] The variance between the predicted withdrawal response time and the actual response time is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal candidate solution;

[0052] Update the particle position, the expression is:

[0053]

[0054] The current number of iterations is t and the maximum number of iterations is t max , the random numbers from 0 to 1 are r2 and r3, and the position vector of the optimal solution for the tth iteration is The position vector of the wth particle at the tth iteration is The updated position vector of the wth particle in the t+1th iteration is

[0055] Update the particle position according to the attack and obtain the attack position. The expression is:

[0056]

[0057] The attack position vector of the wth particle in the t+1th iteration is The updated position vector of the wth particle in the tth iteration is The updated position vector of the random particle in the tth iteration is The spiral coefficient is ω, and the random number between -1 and 1 is v;

[0058] The crossover operator is used to perform crossover transformation on the particle population, and the expression is:

[0059]

[0060] The random number from 0 to 1 is r4, and the random crossover operator is The position of the best candidate solution in the zth generation population is kz , the position of the best candidate solution in the z+1 generation population is k z+1 , the position of other random particles in the zth generation population is k1 z , the position of other random particles in the z+1 generation population is k1 z+1 ;

[0061] Continue to iterate until the fitness value is less than 0.112, otherwise update the particle population and recalculate the fitness function.

[0062] In the second aspect, a quick cash withdrawal system for binding a bank card to a third-party payment platform includes:

[0063] Data collection module: used to collect withdrawal data and bank data from third-party payment platforms, and pre-process the withdrawal data and related data; the withdrawal data includes third-party payment platform information data and user withdrawal data;

[0064] Assessment and verification module: used to perform multimodal risk assessment on the withdrawal data to obtain risk level, and to perform multimodal identity verification based on bank data using biometrics to obtain first verification data;

[0065] Verification construction module: used for performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data and the second verification data;

[0066] Optimization output module: used to build a third-party payment platform quick withdrawal model according to the verification function, optimize the third-party payment platform quick withdrawal model according to the response error, and output the withdrawal result.

[0067] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0068] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0069] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0070] The beneficial effects of the present invention are:

[0071] The present invention is a method and system for quickly withdrawing cash by binding a bank card to a third-party payment platform. Compared with the prior art, the present invention has the following technical effects:

[0072] The present invention can improve the accuracy of user intention recognition in multi-dimensional human-computer interaction scenarios through preprocessing, obtaining satisfactory deviation, obtaining fuzzy intention, obtaining corrected intention, obtaining target intention and model building steps, thereby improving the precision of user intention recognition in multi-dimensional human-computer interaction scenarios, and optimizing user intention recognition in multi-dimensional human-computer interaction scenarios, which can greatly save resources and improve work efficiency. It can realize automatic recognition of user intention in multi-dimensional human-computer interaction scenarios, and perform intention correction and knowledge enhancement on user intention recognition in multi-dimensional human-computer interaction scenarios in real time, which is of great significance to user intention recognition in multi-dimensional human-computer interaction scenarios, can adapt to user intention recognition in multi-dimensional human-computer interaction scenarios with different standards and user intention recognition needs in different multi-dimensional human-computer interaction scenarios, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flowchart of the steps of a method for quickly withdrawing cash by binding a bank card to a third-party payment platform according to the present invention;

[0074] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0075] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0076] A method and system for quickly withdrawing cash by binding a bank card to a third-party payment platform of the present invention comprises the following steps:

[0077] like Figure 1 As shown, in this embodiment, the following steps are included:

[0078] Collecting withdrawal data and bank data from a third-party payment platform, and pre-processing the withdrawal data and related information; the withdrawal data includes third-party payment platform information data and user withdrawal data;

[0079] In the actual assessment, the third-party payment platform information data includes transaction ID, transaction time, transaction amount, payment platform name, and merchant category; user withdrawal data includes user ID, name, account balance, withdrawal amount, withdrawal time, IP address, and device information;

[0080] Bank data includes account information and user identity information; account information includes account ID, user ID, account type, account balance, and transaction records; user identity information includes user ID, name, ID number, mobile phone number, and biometric information;

[0081] User 1 and User 2 are taken as the research objects. User 1’s transaction ID is T123, transaction time is 2024-12-10 14:30, transaction amount is 500 yuan, payment platform name is Alipay, and merchant category is e-commerce; User 1’s withdrawal data: User ID is U101, name is Zhang San, account balance is 1,000 yuan, withdrawal amount is 300 yuan, withdrawal time is 2024-12-10 14:45, IP address is 192.168.1.3, and device information is iPhone 13;

[0082] User 2's third-party payment platform information data: transaction ID is T789, transaction time is 2024-12-10 15:00, transaction amount is 1,500 yuan, payment platform name is Caifutong, merchant category is offline retail; User 2's withdrawal data: user ID is U101, name is Li Si, account balance is 2,000 yuan, withdrawal amount is 1,800 yuan, withdrawal time is 2024-12-10 15:05, IP address is 192.168.1.3, device information is Samsung S21;

[0083] Performing a multimodal risk assessment on the withdrawal data to obtain a risk level, and using biometrics to perform a multimodal identity verification based on bank data to obtain first verification data;

[0084] In the actual evaluation, the risk level of user 1 is 0.3, and the first verification data of user 1 is: success; the risk level of user 2 is 0.7, and the first verification data of user 2 is: failure;

[0085] Performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data, and the second verification data;

[0086] In the actual evaluation, the second verification data of user 1 is: success; the second verification data of user 2 is: failure;

[0087] Constructing a third-party payment platform quick withdrawal model according to the verification function, optimizing the third-party payment platform quick withdrawal model according to the response error, and outputting the withdrawal result;

[0088] In the actual evaluation, User 1 successfully withdrew 300 yuan, and the withdrawal request was successfully processed at 2024-12-10 14:46; User 2 failed to withdraw money.

[0089] In this embodiment, the method of performing multimodal risk assessment on the withdrawal data to obtain a risk degree includes:

[0090] The evaluation indicators are the transaction amount, transaction location, transaction time, transaction type, fund destination and transaction location image of the withdrawal data;

[0091] Based on the historical data of the evaluation indicators, a deep learning algorithm is used to extract the user's behavior habits as a reference object; the evaluation indicators are subjected to real-time feature extraction to obtain evaluation features and evaluation feature values;

[0092] Normalize the evaluation features and evaluation feature values, and calculate the difference of the evaluation features based on the reference object:

[0093]

[0094] The difference of the i-th evaluation feature is The i-th evaluation feature is b i , evaluate feature b i The reference object is b o , the error constant is σ, the optimization coefficient is ζ, and the evaluation feature b i and reference object b o The matching degree is

[0095]

[0096] Calculate the risk of the assessment feature:

[0097]

[0098] The risk of the i-th evaluation feature is K c , the scale factor is λ, and the adjustment coefficient of the i-th evaluation feature is δ i , the correlation between the i-th assessment feature and risk is The number of evaluation features is m c , evaluate feature b i The characteristic value of i |;

[0099] Calculate the risk association of the assessed features:

[0100]

[0101] The risk association of the i-th evaluation feature is θ i , difference The probability of the risk of cash withdrawal occurring at the same time is

[0102] The evaluation features with a risk correlation greater than 0.3061 are taken as key features, and the risk is calculated based on the key features:

[0103]

[0104] The risk of the cth evaluation is K c , the number of key features is m c1 .

[0105] In this embodiment, the method of obtaining first verification data by performing multimodal identity verification based on bank data using biometrics includes:

[0106] Input the fingerprint and face image of the withdrawal data into a multimodal biometric recognition model based on deep learning to obtain multimodal features;

[0107] Among them, the multimodal biometric recognition algorithm: by using lightweight means such as group convolution and channel shuffling, the ghost network and shuffle network lightweight network models are structurally improved; the CBAM attention mechanism is introduced, and the improved PLAM mechanism is proposed to form a multimodal biometric recognition model;

[0108] Obtain the user's identity information database and calculate the matching degree between the user features in the identity information database and the multimodal features:

[0109]

[0110] The ath genetic coefficient is ζ a , the ath user feature is u a , the a-th multimodal feature is d a , the matching degree of the sth identity verification is E s , the authentication is correct cr, user characteristics u a The joint probability distribution of the correct authentication cr is p(u a ,cr), user feature u a The marginal distribution of a ), the marginal distribution of the correct identity authentication cr is p(cr), the decision function is γ(·), and the user feature u a and multimodal features d a The cosine similarity is The number of matching features is N1, and the value of cosine similarity is

[0111] The output of the matching degree greater than 0.658 is considered as identity verification success, otherwise it is considered as identity verification failure, where identity verification success is 1 and identity verification failure is 0;

[0112] The identity verification result is output as the first verification data.

[0113] In this embodiment, the method of constructing a verification function according to the risk level, the first verification data and the second verification data includes:

[0114]

[0115] The verification function is The first verification data is H1, the second verification data is H2, and the risk level of the cth evaluation is K c ;

[0116] When the risk level is greater than 0.682, a danger warning is triggered and the account is locked, and a password verification is performed. The password verification includes a danger warning option and a no-risk option.

[0117] When the verification function is 0, it means that the withdrawal is not possible; when the verification function is not 0, if the verification risk is within the safe range, the withdrawal is allowed, otherwise the withdrawal is not possible;

[0118] In actual evaluation, the secret phrase is set by the user.

[0119] In this embodiment, the method for constructing a third-party payment platform quick withdrawal model according to the verification function includes:

[0120] According to the predicted withdrawal result obtained by the verification function, the minimum loss function of the actual withdrawal result and the predicted withdrawal result is used as the objective function of the third-party payment platform's fast withdrawal model;

[0121] The third-party payment platform's fast cash withdrawal model includes density clustering algorithm, autoencoder, and neural network algorithm;

[0122] The density clustering algorithm identifies the high-density and low-density areas of the input data, and regards the points in the high-density area as normal data, and the points in the low-density area or on the boundary as abnormal data;

[0123] The autoencoder encodes normal data into a low-dimensional hidden layer space through unsupervised learning, and then reconstructs it back to the original input space, learns the compressed representation of normal data, and uses the compressed representation as the extraction feature;

[0124] The neural network algorithm learns withdrawal characteristics and patterns, performs real-time classification and risk assessment based on withdrawal requests, and uses verification functions based on risk assessment to make fast withdrawal decisions on third-party payment platforms.

[0125] In this embodiment, the method for optimizing the third-party payment platform quick cash withdrawal model according to the response error includes:

[0126] The swarm optimization algorithm is introduced to perform chaotic mapping on the particle population. The expression is:

[0127] k w+1 =k w +β·(2r1-1)·(1-k w )

[0128] The chaotic mapping population sequence of the w+1th particle is k a+1 , the initial population sequence of the w+1th particle is kw , the chaotic variation factor is β, and the random number from 0 to 1 is r1;

[0129] The variance between the predicted withdrawal response time and the actual response time is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal candidate solution;

[0130] Update the particle position, the expression is:

[0131]

[0132] The current number of iterations is t and the maximum number of iterations is t max , the random numbers from 0 to 1 are r2 and r3, and the position vector of the optimal solution for the tth iteration is The position vector of the wth particle at the tth iteration is The updated position vector of the wth particle in the t+1th iteration is

[0133] Update the particle position according to the attack and obtain the attack position. The expression is:

[0134]

[0135] The attack position vector of the wth particle in the t+1th iteration is The updated position vector of the wth particle in the tth iteration is The updated position vector of the random particle in the tth iteration is The spiral coefficient is ω, and the random number between -1 and 1 is v;

[0136] The crossover operator is used to perform crossover transformation on the particle population, and the expression is:

[0137]

[0138] The random number from 0 to 1 is r4, and the random crossover operator is The position of the best candidate solution in the zth generation population is k z , the position of the best candidate solution in the z+1 generation population is k z+1 , the position of other random particles in the zth generation population is k1 z , the position of other random particles in the z+1 generation population is k1 z+1 ;

[0139] Continue to iterate until the fitness value is less than 0.112, otherwise update the particle population and recalculate the fitness function.

[0140] In the second aspect, a quick cash withdrawal system for binding a bank card to a third-party payment platform includes:

[0141] Data collection module: used to collect withdrawal data and bank data from third-party payment platforms, and pre-process the withdrawal data and related data; the withdrawal data includes third-party payment platform information data and user withdrawal data;

[0142] Assessment and verification module: used to perform multimodal risk assessment on the withdrawal data to obtain risk level, and to perform multimodal identity verification based on bank data using biometrics to obtain first verification data;

[0143] Verification construction module: used for performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data and the second verification data;

[0144] Optimization output module: used to build a third-party payment platform quick withdrawal model according to the verification function, optimize the third-party payment platform quick withdrawal model according to the response error, and output the withdrawal result.

[0145] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0146] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0147] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0148] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a fast cash withdrawal device for binding a bank card to a third-party payment platform at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned fast cash withdrawal methods for binding a bank card to a third-party payment platform.

[0149] The above application Figure 1 The embodiment shown discloses a method for fast cash withdrawal of a third-party payment platform bound to a bank card, which can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0150] The electronic device may also perform Figure 1 A fast cash withdrawal method for binding bank cards on a third-party payment platform, and realize Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0151] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple applications, any of the aforementioned methods for quickly withdrawing cash from a bank card bound to a third-party payment platform is executed.

[0152] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt 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 include computer-usable program code.

[0153] The present application 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 the present application. 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 generate 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.

[0154] 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.

[0155] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0156] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0157] 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.

[0158] 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 tape 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.

[0159] 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.

[0160] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on 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.

[0161] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for quickly withdrawing cash by binding a bank card to a third-party payment platform, characterized in that: The following steps are involved: Collecting withdrawal data and bank data from a third-party payment platform, and pre-processing the withdrawal data and related information; the withdrawal data includes third-party payment platform information data and user withdrawal data; Performing a multimodal risk assessment on the withdrawal data to obtain a risk level, and using biometrics to perform a multimodal identity verification based on bank data to obtain first verification data; Performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data, and the second verification data; A third-party payment platform quick withdrawal model is constructed according to the verification function, the third-party payment platform quick withdrawal model is optimized according to the response error, and the withdrawal result is output.

2. According to claim 1, a method for quickly withdrawing cash by binding a bank card to a third-party payment platform, characterized in that: The method of performing multimodal risk assessment on the withdrawal data to obtain a risk degree includes: The evaluation indicators are the transaction amount, transaction location, transaction time, transaction type, fund destination and transaction location image of the withdrawal data; Based on the historical data of the evaluation indicators, a deep learning algorithm is used to extract the user's behavior habits as a reference object; the evaluation indicators are subjected to real-time feature extraction to obtain evaluation features and evaluation feature values; Normalize the evaluation features and evaluation feature values, and calculate the difference of the evaluation features based on the reference object: The difference of the i-th evaluation feature is The i-th evaluation feature is b i , evaluate feature b i The reference object is b o , the error constant is σ, the optimization coefficient is ζ, and the evaluation feature b i and reference object b o The matching degree is Calculate the risk of the assessment feature: The risk of the i-th evaluation feature is K c , the scale factor is λ, and the adjustment coefficient of the i-th evaluation feature is δ i , the correlation between the i-th assessment feature and risk is The number of evaluation features is m c , evaluate feature b i The characteristic value of i |; Calculate the risk association of the assessed features: The risk association of the i-th evaluation feature is θ i , difference The probability of the occurrence of the risk of cash withdrawal at the same time is The evaluation features with a risk correlation greater than 0.3061 are taken as key features, and the risk is calculated based on the key features: The risk of the cth evaluation is K c , the number of key features is m c1 .

3. According to claim 1, a quick withdrawal method for a third-party payment platform bound to a bank card is characterized in that: The method of obtaining first verification data by performing multimodal identity verification based on bank data using biometrics includes: Input the fingerprint and face image of the withdrawal data into a multimodal biometric recognition model based on deep learning to obtain multimodal features; Among them, the multimodal biometric recognition algorithm: by using lightweight means such as group convolution and channel shuffling, the ghost network and shuffle network lightweight network models are structurally improved; the CBAM attention mechanism is introduced, and the improved PLAM mechanism is proposed to form a multimodal biometric recognition model; Obtain the user's identity information database and calculate the matching degree between the user features in the identity information database and the multimodal features: The ath genetic coefficient is ζ a , the ath user feature is u a , the a-th multimodal feature is d a , the matching degree of the sth identity verification is E s , the authentication is correct cr, user characteristics u a The joint probability distribution of the correct authentication cr is p(u a ,cr), user feature u a The marginal distribution of a ), the marginal distribution of the correct identity authentication cr is p(cr), the decision function is γ(·), and the user feature u a and multimodal features d a The cosine similarity is The number of matching features is N1, and the value of cosine similarity is The output of the matching degree greater than 0.658 is considered as identity verification success, otherwise it is considered as identity verification failure, where identity verification success is 1 and identity verification failure is 0; The identity verification result is output as the first verification data.

4. According to claim 1, a method for quickly withdrawing cash by binding a bank card to a third-party payment platform, characterized in that: The method for constructing a verification function according to the risk degree, the first verification data and the second verification data comprises: The verification function is The first verification data is H1, the second verification data is H2, and the risk level of the cth evaluation is K c ; When the risk level is greater than 0.682, a danger warning is triggered and the account is locked, and a password verification is performed. The password verification includes a danger warning option and a no-risk option. When the verification function is 0, it means that withdrawals are not possible; when the verification function is not 0, if the verification risk is within a safe range, withdrawals can be made, otherwise withdrawals cannot be made.

5. According to claim 1, a quick withdrawal method for a third-party payment platform bound to a bank card is characterized in that: The method for constructing a third-party payment platform quick withdrawal model according to the verification function includes: According to the predicted withdrawal result obtained by the verification function, the minimum loss function of the actual withdrawal result and the predicted withdrawal result is used as the objective function of the third-party payment platform's fast withdrawal model; The third-party payment platform's fast cash withdrawal model includes density clustering algorithm, autoencoder, and neural network algorithm; The density clustering algorithm identifies the high-density and low-density areas of the input data, and regards the points in the high-density area as normal data, and the points in the low-density area or on the boundary as abnormal data; The autoencoder encodes normal data into a low-dimensional hidden layer space through unsupervised learning, and then reconstructs it back to the original input space, learns the compressed representation of normal data, and uses the compressed representation as the extraction feature; The neural network algorithm learns withdrawal characteristics and patterns, performs real-time classification and risk assessment based on withdrawal requests, and uses verification functions based on risk assessment to make fast withdrawal decisions on third-party payment platforms.

6. According to claim 1, a method for quickly withdrawing cash by binding a bank card to a third-party payment platform, characterized in that: The method for optimizing the third-party payment platform quick cash withdrawal model according to the response error includes: The swarm optimization algorithm is introduced to perform chaotic mapping on the particle population. The expression is: k w+1 =k w +β·(2r1-1)·(1-k w ) The chaotic mapping population sequence of the w+1th particle is k a+1 , the initial population sequence of the w+1th particle is k w , the chaotic variation factor is β, and the random number from 0 to 1 is r1; The variance between the predicted withdrawal response time and the actual response time is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal candidate solution; Update the particle position, the expression is: The current number of iterations is t and the maximum number of iterations is t max , the random numbers from 0 to 1 are r2 and r3, and the position vector of the optimal solution for the tth iteration is The position vector of the wth particle at the tth iteration is The updated position vector of the wth particle in the t+1th iteration is Update the particle position according to the attack and obtain the attack position. The expression is: The attack position vector of the wth particle in the t+1th iteration is The updated position vector of the wth particle in the tth iteration is The updated position vector of the random particle in the tth iteration is The spiral coefficient is ω, and the random number between -1 and 1 is v; The crossover operator is used to perform crossover transformation on the particle population, and the expression is: The random number from 0 to 1 is r4, and the random crossover operator is The position of the best candidate solution in the zth generation population is k z , the position of the best candidate solution in the z+1 generation population is k z+1 , the position of other random particles in the zth generation population is k1 z , the position of other random particles in the z+1 generation population is k1 z+1 ; Continue to iterate until the fitness value is less than 0.112, otherwise update the particle population and recalculate the fitness function.

7. A fast cash withdrawal system for binding a bank card to a third-party payment platform, used to execute the method according to any one of claims 1 to 6, characterized in that: include: Data collection module: used to collect withdrawal data and bank data from third-party payment platforms, and pre-process the withdrawal data and related data; The withdrawal data includes third-party payment platform information data and user withdrawal data; Assessment and verification module: used to perform multimodal risk assessment on the withdrawal data to obtain risk level, and to perform multimodal identity verification based on bank data using biometrics to obtain first verification data; Verification construction module: used for performing account amount verification according to the bank data to obtain second verification data, and constructing a verification function according to the risk level, the first verification data and the second verification data; Optimization output module: used to build a third-party payment platform quick withdrawal model according to the verification function, optimize the third-party payment platform quick withdrawal model according to the response error, and output the withdrawal result.

8. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute the method according to any one of claims 1 to 6.

Images