CN112613871A

CN112613871A - Payment mode recommendation method based on big data and block chain and cloud computing server

Info

Publication number: CN112613871A
Application number: CN202011494657.XA
Authority: CN
Inventors: 黄蕾
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-12-17
Filing date: 2020-12-17
Publication date: 2021-04-06

Abstract

According to the payment mode recommendation method and the cloud computing server based on the big data and the block chain, the generated data network analysis model, the data security protection model and the payment recommendation permission information of the payment service terminal can be stored in the preset database in advance. When the payment mode recommendation is carried out, payment recommendation permission information of the corresponding model can be respectively determined from a preset database according to the payment data transmission information and the data interface updating information of the recommended payment service terminal, and when the recommended payment service terminal is judged to pass the recommendation verification, the recommended payment mode of the recommended payment service terminal can be determined based on the searched corresponding model. Therefore, the payment data transmission information and the data interface updating information of the payment service terminal can be analyzed, the recommended payment mode of the payment service terminal is intelligently determined, a more appropriate payment mode is quickly and accurately recommended to a user, and the extra time cost increased in the payment process is reduced.

Description

Payment mode recommendation method based on big data and block chain and cloud computing server

Technical Field

The application relates to the technical field of big data analysis and block chain payment, in particular to a payment mode recommendation method based on big data and a block chain and a cloud computing server.

Background

With the development of a Blockchain (Blockchain) technology, scenes applied by the Blockchain become more and more extensive, wherein a payment scene is one of main application scenes of the Blockchain.

A conventional payment scenario is shown in fig. 1, where a payment service between a service party X and a service party Y needs to use an intermediate platform Z, which may be limited by various factors, resulting in poor real-time mobility and high commission.

The payment scenario of the blockchain payment is shown in fig. 2, and the service party X and the service party Y can directly perform payment service interaction, so that intermediate links are reduced, the payment cost can be reduced, and the timeliness of the payment service is improved.

With the continuous expansion of the online payment service, today's payment service can be realized by a plurality of payment methods, however, in practical application, a user may spend a lot of time in selecting a payment method, which may increase the extra time cost in the payment process.

Disclosure of Invention

The first aspect of the application discloses a payment mode recommendation method based on big data and a block chain, which comprises the following steps: acquiring payment data transmission information and data interface updating information of a payment service terminal during online payment; respectively training a data network analysis model and a data safety protection model by using the payment data transmission information and the data interface updating information, and correspondingly storing the data network analysis model, the data safety protection model and payment recommendation permission information of the payment service terminal into a preset database; when the payment mode recommendation is performed on any payment service terminal, acquiring payment data transmission information and data interface updating information of the recommended payment service terminal, finding a data network analysis model matched with the acquired payment data transmission information from a preset database, acquiring first payment recommendation permission information corresponding to the found data network analysis model from the preset database, finding a data security protection model matched with the acquired data interface updating information from the preset database, and acquiring second payment recommendation permission information corresponding to the found data security protection model from the preset database; comparing whether the first payment recommendation permission information and the second payment recommendation permission information are the same or not, and verifying whether the recommended payment service terminal passes recommendation verification or not according to a comparison result; and when the recommended payment service terminal passes the recommendation verification, determining a recommended payment mode of the recommended payment service terminal according to the found data network analysis model and the found data security protection model.

A second aspect of the present application discloses a cloud computing server, comprising a processing engine, a network module, and a memory; the processing engine and the memory communicate via the network module, and the processing engine reads the computer program from the memory and runs it to perform the method of the first aspect.

Compared with the prior art, the payment mode recommendation method based on big data and the block chain and the cloud computing server provided by the embodiment of the invention have the following technical effects: the data network analysis model and the data safety protection model which are correspondingly generated when the payment service terminal carries out online payment and the payment recommendation permission information of the payment service terminal can be correspondingly stored in the preset database in advance, so that decision and analysis basis can be provided for payment mode recommendation. In addition, when the payment mode is recommended to the recommended payment service terminal, the payment recommendation permission information of the corresponding model can be respectively determined from the preset database according to the payment data transmission information and the data interface updating information of the recommended payment service terminal, so that the recommended payment mode of the recommended payment service terminal can be determined based on the searched corresponding model when the recommended payment service terminal is judged to pass the recommendation verification. Therefore, the payment data transmission information and the data interface updating information of the payment service terminal can be analyzed, so that the recommended payment mode of the payment service terminal can be intelligently determined, a more appropriate payment mode can be quickly and accurately recommended to a user, and the extra time cost in the payment process is reduced.

In the description that follows, additional features will be set forth, in part, in the description. These features will be in part apparent to those skilled in the art upon examination of the following and the accompanying drawings, or may be learned by production or use. The features of the present application may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations particularly pointed out in the detailed examples that follow.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

The methods, systems, and/or processes of the figures are further described in accordance with the exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments in which reference numerals represent similar mechanisms throughout the various views of the drawings.

Fig. 1 is a schematic diagram of a conventional payment scenario.

Fig. 2 is a schematic diagram of a payment scenario for blockchain payments.

FIG. 3 is a block diagram of an exemplary big data and blockchain based payment method recommendation system according to some embodiments of the invention.

Fig. 4 is a schematic diagram illustrating hardware and software components in an exemplary cloud computing server, according to some embodiments of the invention.

FIG. 5 is a flow diagram illustrating an exemplary big data and blockchain based payment method recommendation method and/or process according to some embodiments of the invention.

Detailed Description

The inventor has found that when the user selects the payment method, the selected payment method may be unreasonable, for example, if the user selects the password payment method, but suddenly forgets the password, the user may select the face-brushing payment method at that time, which increases the extra time cost in the payment process. In order to solve the technical problem, the inventor innovatively provides a payment method recommendation method based on big data and a block chain and a cloud computing server, and can analyze payment data transmission information and data interface updating information of a payment service terminal so as to intelligently determine the recommended payment method of the payment service terminal, so that a more appropriate payment method can be quickly and accurately recommended to a user, and extra time cost in a payment process is reduced.

In order to better understand the technical solutions of the present invention, the following detailed descriptions of the technical solutions of the present invention are provided with the accompanying drawings and the specific embodiments, and it should be understood that the specific features in the embodiments and the examples of the present invention are the detailed descriptions of the technical solutions of the present invention, and are not limitations of the technical solutions of the present invention, and the technical features in the embodiments and the examples of the present invention may be combined with each other without conflict.

In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant guidance. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, systems, compositions, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the invention.

These and other features, functions, methods of execution, and combination of functions and elements of related elements in the structure and economies of manufacture disclosed in the present application may become more apparent upon consideration of the following description with reference to the accompanying drawings, all of which form a part of this application. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. It should be understood that the drawings are not to scale.

Flowcharts are used herein to illustrate the implementations performed by systems according to embodiments of the present application. It should be expressly understood that the processes performed by the flowcharts may be performed out of order. Rather, these implementations may be performed in the reverse order or simultaneously. In addition, at least one other implementation may be added to the flowchart. One or more implementations may be deleted from the flowchart.

Fig. 3 is a block diagram illustrating an exemplary big data and blockchain based payment method recommendation system 300 according to some embodiments of the present invention, where the big data and blockchain based payment method recommendation system 300 may include a cloud computing server 100 and a payment service terminal 200.

In some embodiments, as shown in fig. 4, the cloud computing server 100 may include a processing engine 110, a network module 120, and a memory 130, the processing engine 110 and the memory 130 communicating through the network module 120.

Processing engine 110 may process the relevant information and/or data to perform one or more of the functions described herein. For example, in some embodiments, processing engine 110 may include at least one processing engine (e.g., a single core processing engine or a multi-core processor). By way of example only, the Processing engine 110 may include a Central Processing Unit (CPU), an Application-Specific Integrated Circuit (ASIC), an Application-Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

Network module 120 may facilitate the exchange of information and/or data. In some embodiments, the network module 120 may be any type of wired or wireless network or combination thereof. Merely by way of example, the Network module 120 may include a cable Network, a wired Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a Wireless personal Area Network, a Near Field Communication (NFC) Network, and the like, or any combination thereof. In some embodiments, the network module 120 may include at least one network access point. For example, the network module 120 may include wired or wireless network access points, such as base stations and/or network access points.

The Memory 130 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 130 is used for storing a program, and the processing engine 110 executes the program after receiving the execution instruction.

It is to be understood that the configuration shown in fig. 4 is merely illustrative, and that the cloud computing server 100 may also include more or fewer components than shown in fig. 4, or have a different configuration than shown in fig. 4. The components shown in fig. 4 may be implemented in hardware, software, or a combination thereof.

Fig. 5 is a flowchart illustrating an exemplary big data and blockchain based payment method recommendation method and/or process according to some embodiments of the present invention, where the big data and blockchain based payment method recommendation method is applied to the cloud computing server 100 in fig. 3, and may specifically include the following steps S51 to S55.

Step S51, obtaining payment data transmission information and data interface update information when the payment service terminal performs online payment. For example, the payment service terminal may be a smart phone, a tablet computer, a notebook computer, or other portable terminals, or may be an enterprise-level service terminal, and the like, which is not limited herein. Online payment may be understood as online payment or network payment. The payment data transmission information can be used for representing the transmission condition of service data between the payment service terminal and a counterparty (other service terminals performing payment service interaction with the payment service terminal) when performing online payment, and the data interface updating information is used for representing the opening or closing condition of a data access interface corresponding to the payment service terminal in the payment service process.

Step S52, respectively training a data network analysis model and a data security protection model by using the payment data transmission information and the data interface update information, and correspondingly storing the data network analysis model, the data security protection model and the payment recommendation permission information of the payment service terminal in a preset database. For example, a data network analysis model is used to analyze network data of the payment network, which may be a neural network model, and a data security protection model is used to detect and protect the security of the data access interfaces of the payment service terminals, such as turning off or on certain data access interfaces at appropriate times. The payment recommendation permission information is generated after the cloud computing server performs authorization authentication with the payment service terminal in advance, and is used for representing that the payment service terminal allows the cloud computing server to recommend a payment mode to the payment service terminal, and the preset database can be a database communicated with the cloud computing server. It is understood that the number of the data network analysis model and the number of the data security protection model stored in the preset database may be multiple.

Step S53, when a payment mode is recommended for any payment service terminal, collecting payment data transmission information and data interface updating information of the recommended payment service terminal, finding a data network analysis model matched with the collected payment data transmission information from the preset database, acquiring first payment recommendation permission information corresponding to the found data network analysis model from the preset database, finding a data security protection model matched with the collected data interface updating information from the preset database, and acquiring second payment recommendation permission information corresponding to the found data security protection model from the preset database. For example, the data network analysis model matching the collected payment data transmission information may be understood as a payment portrait matching between the data network analysis model and the collected payment data transmission information, the payment portrait includes but is not limited to a payment region portrait (whether cross-border payment is performed), a payment time period, a payment amount scale, and the like, and the first payment recommendation permission information corresponding to the found data network analysis model may carry a model tag corresponding to the found data network analysis model. Similarly, the data security protection model matched with the collected data interface update information may be understood as a payment portrait matching between the data security protection model and the collected data interface update information, and the second payment recommendation permission information corresponding to the found data security protection model may carry a model tag corresponding to the found data security protection model.

Step S54, comparing whether the first payment recommendation permission information and the second payment recommendation permission information are the same, and verifying whether the recommended payment service terminal passes recommendation verification according to the comparison result. For example, by verifying whether the recommended payment service terminal passes the recommendation verification, the payment service security of the recommended payment service terminal can be ensured before the payment method recommendation is performed.

Step S55, when the recommended payment service terminal passes the recommendation verification, determining the recommended payment mode of the recommended payment service terminal according to the found data network analysis model and the found data security protection model. For example, the recommended payment methods include, but are not limited to, password payment methods, fingerprint payment methods, and face-brushing payment methods.

It can be understood that, by implementing the above steps S51-S55, the data network analysis model and the data security protection model correspondingly generated when the payment service terminal performs online payment, and the payment recommendation permission information of the payment service terminal can be correspondingly stored in the preset database in advance, so that a decision and analysis basis can be provided for the payment mode recommendation. In addition, when the payment mode is recommended to the recommended payment service terminal, the payment recommendation permission information of the corresponding model can be respectively determined from the preset database according to the payment data transmission information and the data interface updating information of the recommended payment service terminal, so that the recommended payment mode of the recommended payment service terminal can be determined based on the searched corresponding model when the recommended payment service terminal is judged to pass the recommendation verification. Therefore, the payment data transmission information and the data interface updating information of the payment service terminal can be analyzed, so that the recommended payment mode of the payment service terminal can be intelligently determined, a more appropriate payment mode can be quickly and accurately recommended to a user, and the extra time cost in the payment process is reduced.

In the following, some alternative embodiments will be described, which should be understood as examples and not as technical features essential for implementing the present solution.

In a further possible example, the verifying the recommended payment service terminal according to the comparison result described in step S54 includes: when the comparison result is that the first payment recommendation permission information is the same as the second payment recommendation permission information, determining that the recommended payment service terminal passes recommendation verification; and when the comparison result shows that the first payment recommendation permission information is different from the second payment recommendation permission information, determining that the recommended payment service terminal does not pass recommendation verification. For example, the first payment recommendation permission information and the second payment recommendation permission information are the same, that is, the first payment recommendation permission information and the second payment recommendation permission information are in one-to-one correspondence or at least partially in correspondence with each other on a plurality of payment images, for example, the first payment recommendation permission information and the second payment recommendation permission information respectively include 10 payment images, and if 6 identical payment images exist between the first payment recommendation permission information and the second payment recommendation permission information, it may be determined that the recommended payment service terminal passes recommendation verification.

In a further possible example, the step S51 of obtaining the payment data transmission information and the data interface update information when the payment service terminal performs online payment further includes the step S510 of: and acquiring service processing information uploaded by the payment service terminal during online payment.

In a further possible example, the step S51 of storing the data network analysis model, the data security protection model and the payment recommendation permission information of the payment service terminal in a preset database correspondingly includes: and correspondingly storing a data network analysis model, a data security protection model, the service processing information and payment recommendation permission information of the payment service terminal into a preset database.

It is understood that the verification of whether the recommended payment service terminal passes the recommendation verification according to the comparison result described in the step S54 includes the following steps S541 and S542 on the basis of the step S510.

Step S541, when the comparison result is that the first payment recommendation permission information is the same as the second payment recommendation permission information, acquiring service processing information corresponding to the first payment recommendation permission information from the preset database, and verifying whether the recommended payment service terminal passes recommendation verification by using the acquired service processing information.

And step S542, when the comparison result is that the first payment recommendation permission information is different from the second payment recommendation permission information, determining that the recommended payment service terminal does not pass the recommendation verification, or respectively acquiring service processing information corresponding to the first payment recommendation permission information and the second payment recommendation permission information, and verifying whether the recommended payment service terminal passes the recommendation verification by using the service processing information corresponding to the first payment recommendation permission information and the service processing information corresponding to the second payment recommendation permission information.

Still further, the verifying whether the recommended payment service terminal passes the recommendation verification by using the acquired service processing information described in step S541 may include the following two embodiments.

In the first implementation mode, interface update state analysis is performed on the collected data interface update information of the recommended payment service terminal, whether the result obtained by the interface update state analysis is consistent with the obtained service processing information or not is compared, if so, the recommended payment service terminal is determined to pass recommendation verification, and otherwise, the recommended payment service terminal is determined not to pass recommendation verification.

In the second implementation mode, the payment service matching degree is calculated for the collected data interface updating information of the recommended payment service terminal and the acquired service processing information; and identifying whether the corresponding matching degree calculation result is greater than a set threshold value, if so, determining that the recommended payment service terminal passes the recommendation verification, otherwise, determining that the recommended payment service terminal does not pass the recommendation verification.

Still further, the verifying whether the recommended payment service terminal passes the recommendation verification by using the service processing information corresponding to the first payment recommendation permission information and the service processing information corresponding to the second payment recommendation permission information described in step S542 includes steps S5421 to S5424 as follows.

Step S5421, payment service matching degree calculation is carried out on the collected data interface updating information of the recommended payment service terminal and the service processing information corresponding to the first payment recommendation permission information, and the first payment service matching degree is obtained. For example, the value range of the payment service matching degree may be 0 to 1, the calculation of the payment service matching degree may be implemented by a preset calculation thread, and the functional adjustment and debugging of the calculation thread are not described herein again.

Step S5422, performing payment service matching degree calculation on the collected data interface update information of the recommended payment service terminal and the service processing information corresponding to the second payment recommendation permission information, to obtain a second payment service matching degree.

Step S5423, if the matching degree of the first payment service is greater than the matching degree of the second payment service, and the matching degree of the first payment service is also greater than the set threshold, it is determined that the recommended payment service terminal is the payment service terminal corresponding to the first payment recommendation permission information, and it is determined that the payment service terminal corresponding to the first payment recommendation permission information passes the recommendation verification, otherwise, it is determined that the payment service terminal corresponding to the first payment recommendation permission information fails the recommendation verification. For example, the set threshold may be adjusted and modified according to actual conditions, and is not limited herein.

Step S5424, if the second payment service matching degree is greater than the first payment service matching degree and the second payment service matching degree is also greater than the set threshold, determining that the recommended payment service terminal is the payment service terminal corresponding to the second payment recommendation permission information, and determining that the payment service terminal corresponding to the second payment recommendation permission information passes the recommendation verification, otherwise, determining that the payment service terminal corresponding to the second payment recommendation permission information does not pass the recommendation verification.

The above-mentioned preferred verification embodiments are parallel embodiments, and these parallel embodiments may be implemented by selecting one of them according to actual situations, and are not limited herein. It can be understood that, by the parallel embodiment, the selectivity of recommendation verification of the payment service terminal is increased, so that a suitable recommendation verification embodiment can be selected to be implemented in different scenes.

In some possible embodiments, the determining the recommended payment method of the recommended payment service terminal according to the found data network analysis model and the found data security model described in step S55 may include steps S551 to S555 as follows.

And S551, acquiring the current payment environment data of the recommended payment service terminal according to the found data network analysis model and the found data security protection model. For example, the current payment environment data may be network environment data or terminal operation state data, and the like, which is not limited herein.

Step S552, obtaining effective behavior data of the payment service behavior in the current payment environment data. For example, the valid behavior data refers to behavior data that can be recognized and processed by the recommended payment service terminal.

Step S553, extracting the behavior data feature of the valid behavior data. For example, the behavioral data features may be feature vectors or other forms of features.

And step S554, classifying the behavior data characteristics of the effective behavior data by password payment, fingerprint payment and face swiping payment according to a password payment classification model, a fingerprint payment classification model and a face swiping payment classification model in sequence to obtain the password payment behavior characteristics, the fingerprint payment behavior characteristics and the face swiping payment behavior characteristics of the payment business behavior. For example, the classification model may be a classifier.

And step 555, determining the recommended payment mode of the recommended payment service terminal according to the password payment behavior characteristic, the fingerprint payment behavior characteristic and the face brushing payment behavior characteristic.

By implementing the steps S551 to S555, the current payment environment data of the recommended payment service terminal can be analyzed, so that behavior data characteristics of effective behavior data are determined, payment classification of the behavior data characteristics is realized based on different classification models, and different payment behavior characteristics are determined, so that a recommended payment mode suitable for the current payment state of the recommended payment service terminal can be accurately determined according to different payment behavior characteristics, and further, extra time cost added in the payment process is reduced.

In some other embodiments that can be combined, before the step of obtaining the valid behavior data of the payment service behavior in the current payment environment data described in the step S552, a step S60 is further included: and sequentially establishing a password payment classification model, a fingerprint payment classification model and a face brushing payment classification model.

Further, the step S60 of establishing the password payment classification model, the fingerprint payment classification model and the face-brushing payment classification model in sequence includes the following steps S61-S64.

Step S61, a training sample set is obtained. For example, the training sample set may be obtained according to historical payment recommendation records, which is not described herein.

Step S62, obtaining effective behavior data of each current payment environment data in the training sample set.

And step S63, extracting the behavior data characteristics of each effective behavior data.

And step S64, sequentially establishing a password payment classification model, a fingerprint payment classification model and a face brushing payment classification model according to the behavior data characteristics of the effective behavior data of all the current payment environment data in the training sample set.

Therefore, by training different classification models, the classification accuracy and the reliability of the classification models in actual application can be ensured, and a reliable decision basis is provided for recommendation of payment modes.

In a further embodiment, in step S64, establishing a password payment classification model according to the behavior data features in the training sample set includes: dividing the training sample set into pure letter password payment training samples, pure digital password payment training samples and mixed password payment training samples according to three types of sample tags of the pure letter password payment sample tags, the pure digital password payment sample tags and the mixed password payment sample tags; and respectively training the behavior data characteristics of the pure letter password payment training sample, the behavior data characteristics of the pure digital password payment training sample and the behavior data characteristics of the mixed password payment training sample to obtain three classification models corresponding to the password payment, wherein the three classification models respectively correspond to a sample label of the pure letter password payment, a sample label of the pure digital password payment and a sample label of the mixed password payment.

In a further embodiment, in step S64, establishing a fingerprint payment classification model according to the behavior data features of the training sample set includes: dividing the pure letter password payment training sample, the pure digital password payment training sample and the mixed password payment training sample again according to fingerprint payment statistical information to obtain the fingerprint payment sample; respectively training the behavior data characteristics of the fingerprint payment samples to obtain two classification models corresponding to the fingerprint payment samples, wherein the two classification models respectively correspond to point contact payment and press type payment;

in a further embodiment, in step S64, a face-brushing payment classification model is established according to the behavior data features of the training sample set, which includes steps S641-S644.

Step S641 is to divide the associated behavior data features corresponding to the fingerprint payment samples again according to the face-brushing payment division index, so as to obtain a left half face region training sample, a right half face region training sample, an upper half face region training sample, and a lower half face region training sample.

And S642, training the left half face region training sample, the right half face region training sample, the upper half face region training sample and the lower half face region training sample to construct a first-level face brushing payment classification model.

And S643, training the behavior data characteristics of the target face region training sample corresponding to the set face feature recognition degree by taking the set face feature recognition degree as a reference grade under the divided face region training sample to obtain a second-stage face brushing payment classification model.

Step S644 is to train a third-level face brushing payment classification model having a dynamic face region fitting function for each target face region training sample for which a face feature recognition degree is set, under the divided face region training samples.

It can be understood that, in the process of training different classification models, the relevance between different payment samples can be properly considered, so that global model training is realized.

In some other embodiments, the obtaining effective behavior data of payment transaction behavior in the current payment environment data in step S552 includes: detecting payment business behaviors in the current payment environment data, and determining execution parameters of a behavior function of the payment business behaviors; and correcting the original behavior data of the payment business behavior based on the execution parameters of the behavior function of the payment business behavior, and analyzing and setting the payment business behavior in a payment time interval by taking the behavior function of the payment business behavior as a reference to obtain the effective behavior data of the payment business behavior. For example, the related meanings of the behavior function and the behavior parameter can be obtained according to related patent documents (e.g., patents related to user behavior) or the prior art, and are not described herein.

In some further embodiments, the extracting of the behavior data feature of the valid behavior data described in step S553 includes: and extracting the multidimensional description characteristics of the behavior data of the payment business behavior aiming at the effective behavior data of the payment business behavior. For example, behavioral data multidimensional characterization features enable characterization and differentiation of valid behavioral data from multiple feature dimensions. These dimensions include, but are not limited to, the size of the behavior data, the format of the behavior data, the associations of the behavior data, and the like.

In a further embodiment, the step S554 sequentially performs classification of password payment, fingerprint payment and face swiping payment on the behavior data features of the valid behavior data according to a password payment classification model, a fingerprint payment classification model and a face swiping payment classification model to obtain the password payment behavior features, the fingerprint payment behavior features and the face swiping payment behavior features of the payment service behavior, and may include steps S5541 to S5543.

Step S5541, password payment classification is carried out on the behavior data characteristics of the effective behavior data according to a password payment classification model, and password payment behavior characteristics of the payment business behavior are obtained.

Step S5542, selecting a fingerprint payment classification model corresponding to the password payment behavior characteristics of the payment business behavior, and performing fingerprint payment classification on the behavior data characteristics of the effective behavior data according to the fingerprint payment classification model to obtain the fingerprint payment behavior characteristics of the payment business behavior.

Step S5543, selecting a face-brushing payment classification model corresponding to the password payment behavior characteristics and the fingerprint payment behavior characteristics of the payment business behaviors, and performing face-brushing payment classification on the behavior data characteristics of the effective behavior data according to the face-brushing payment classification model to obtain the face-brushing payment behavior characteristics of the payment business behaviors.

On the basis of the step S5543, the inventor finds that, when determining the face-brushing payment behavior feature, it is necessary to consider a dynamic change of a facial region, and to achieve this purpose, selects a face-brushing payment classification model corresponding to the password payment behavior feature and the fingerprint payment behavior feature of the payment business behavior, and performs face-brushing payment classification on the behavior data feature of the valid behavior data according to the face-brushing payment classification model to obtain the face-brushing payment behavior feature of the payment business behavior, which may include step S5543 a-step S5543 d.

Step S5543a, selecting a first-level face-brushing payment classification model corresponding to password payment behavior characteristics and fingerprint payment behavior characteristics of the payment business behavior, and inputting behavior data characteristics of the effective behavior data into the first-level face-brushing payment classification model to obtain first-level face region dynamic characteristics corresponding to the effective behavior data and face behavior characteristic weights corresponding to each first-level face region dynamic characteristic.

Step S5543b, selecting a second-level face brushing payment classification model according to the first-level face region dynamic feature with the largest facial behavior feature weight corresponding to the first-level face region dynamic feature, and inputting the behavior data feature of the valid behavior data into the second-level face brushing payment classification model, to obtain the second-level face region dynamic feature corresponding to the valid behavior data and the facial behavior feature weight corresponding to each second-level face region dynamic feature.

Step S5543c, selecting a third-level face brushing payment classification model according to the second-level facial region dynamic feature with the largest facial behavior feature weight corresponding to each second-level facial region dynamic feature, and inputting the behavior data feature of the valid behavior data into the third-level face brushing payment classification model to obtain a third-level facial region dynamic feature corresponding to the valid behavior data.

Step S5543d, obtaining a face brushing payment behavior feature of the payment business behavior according to the third-level facial region dynamic feature, the facial behavior feature weight corresponding to the first-level facial region dynamic feature, and the facial behavior feature weight corresponding to the second-level facial region dynamic feature.

It should be understood that the meaning of the related art features in the above-mentioned step S5543 a-step S5543d can be determined by the existing feature classification technology, and is not explained herein one by one for reasons of space. It can be understood that, by implementing the above-mentioned steps S5543 a-S5543 d, a multi-level face-brushing payment classification model can be used to analyze dynamic characteristics of a face region when determining the face-brushing payment behavior characteristics, so that dynamic changes of the face region can be considered, and errors in determination of the face-brushing payment behavior characteristics due to normal dynamic changes of the face can be avoided.

In a further embodiment, the step S555 of determining the recommended payment method of the recommended payment service terminal through the password payment behavior characteristic, the fingerprint payment behavior characteristic and the face swiping payment behavior characteristic may include the following steps S5551 to S5555.

Step S5551, according to password payment action characteristic fingerprint payment action characteristic with brush face payment action characteristic, acquire first payment action and merge the characteristic set, wherein, first payment action merges the characteristic set and includes n first payment action and merges the characteristic, and every first payment action merges the characteristic and has m payment action characteristic distribution information, n is the integer that is greater than 1, m is the integer that is greater than 1. For example, the distribution information may be a distribution list or a distribution list, which is not limited herein.

Step S5552 is to generate a first recommended payment behavior feature set according to the first payment behavior fusion feature set, where the first recommended payment behavior feature set includes n first recommended payment behavior features, each first recommended payment behavior feature is obtained by feature screening of the first payment behavior fusion feature, and each first recommended payment behavior feature has m pieces of recommended feature distribution information.

Step S5553, for target payment behavior feature distribution information, determining a first payment formula recommendation index according to the first recommended payment behavior feature set, where the target payment behavior feature distribution information belongs to any one recommended feature distribution information in the m pieces of recommended feature distribution information.

Step S5554, regarding the target payment behavior characteristic distribution information, if a payment security verification condition is satisfied, using the first payment formula recommendation index as the current recommendation index of the recommended payment service terminal corresponding to the target payment behavior characteristic distribution information.

Step S5555, determining the recommended payment mode of the recommended payment service terminal according to the current recommendation index.

On the basis of the step S5555, determining the recommended payment method of the recommended payment service terminal according to the current recommendation index includes: and determining the recommendation index of each candidate payment mode according to the current recommendation index, and determining the candidate payment mode with the maximum recommendation index as the recommended payment mode, wherein the candidate payment modes comprise a password payment mode, a fingerprint payment mode and a face brushing payment mode. For example, the recommended index may be 0 to 1, or 0 to 100, and is not limited herein.

By the design, the password payment behavior characteristic, the fingerprint payment behavior characteristic and the face brushing payment behavior characteristic can be subjected to fusion analysis when the recommended payment mode is determined, so that various user operations of the recommended payment service terminal can be considered as far as possible by the current recommendation index, and accurate recommendation of the recommended payment mode can be achieved.

In an alternative embodiment, after the recommended payment method is determined, a step of performing payment security risk detection on the recommended payment service terminal may be further included, and when it is determined that the recommended payment service terminal does not have a payment security risk, the corresponding recommended payment method is sent to the recommended payment service terminal.

On the basis of the above content, the security risk detection for the recommended payment service terminal may include the content described in the following steps S110 to S150.

Step S110, obtaining payment item statistical information corresponding to each remote payment item in a recommended payment service terminal to obtain i payment item statistical information, wherein the recommended payment service terminal comprises i remote payment items, the remote payment items and the payment item statistical information have a one-to-one correspondence relationship, and i is an integer greater than or equal to 1.

For example, the remote payment transaction may be different remote payment services, such as a transfer service, a pending payment service, a recharge service, and the like, which is not limited herein. The payment statistics information is used for recording and counting information of the remote payment service related to the corresponding remote payment, such as time, place, participant, fund size, etc. of the remote payment service, which is not limited herein.

Step S120, determining i payment security verification information according to the i payment transaction statistical information, where the payment security verification information indicates a security verification record of the remote payment transaction for the recommended payment service terminal in each payment network environment, and the payment security verification information and the remote payment transaction have a one-to-one correspondence relationship.

For example, the security verification records of the remote payment items are different when the recommended payment service terminal is in different payment network environments, and if the recommended payment service terminal is in different payment network environments in the processing process of the same remote payment item, the plurality of security verification records recorded in the payment security verification information are also different.

Step S130, i security verification tags are determined according to the i payment security verification information, wherein the security verification tags and the remote payment items have one-to-one correspondence.

For example, the security verification tag is used to distinguish payment security verification information, and the security verification tag may be divided into many types according to actual situations, which is not limited herein.

Step S140, based on the i security verification tags, obtaining i payment information detection results corresponding to the recommended payment service terminal through a payment information detection model, where the payment information detection results and the remote payment items have a one-to-one correspondence relationship.

For example, the payment information detection model may be a neural network model, and the payment information detection model may be obtained through pre-training. The payment information detection result is used for representing information safety detection results of the recommended payment service terminal in different payment network environments, and the i payment information detection results are different from each other.

And step S150, judging whether the recommended payment service terminal has a payment safety risk or not according to the detection result of the i payment information.

For example, the payment security risk may be a risk that data information of the recommended payment service terminal is stolen or illegally accessed.

It can be understood that, when the above steps S110 to S150 are applied, first, i pieces of payment statistic information of i pieces of remote payment items in the recommended payment service terminal are obtained, then, i pieces of payment security verification information are determined, and further, i pieces of security verification tags are determined, so that i pieces of payment information detection results are obtained through the payment information detection model, and thus, whether the recommended payment service terminal has a payment security risk can be judged through the i pieces of payment information detection results. By the design, when the payment safety risk of the recommended payment service terminal is judged, the safety verification record of the remote payment item in each payment network environment of the recommended payment service terminal is considered, so that the payment safety risk detection of the recommended payment service terminal can be realized from the remote payment item level, the payment network environment level and the data interface verification level, the possible payment safety risk of the recommended payment service terminal can be quickly and accurately found, and the loss of important information and data of the recommended payment service terminal is avoided.

In some possible examples, the obtaining of the i pieces of payment transaction statistical information for the payment transaction statistical information corresponding to each remote payment transaction in the recommended payment service terminal in step S110 may be implemented by the following steps S111 to S113.

Step S111, for a jth remote payment transaction in the recommended payment service terminal, obtaining a payment transaction tag corresponding to the jth remote payment transaction, where the jth remote payment transaction is any one of the i remote payment transactions, and j is an integer greater than 0 and less than or equal to i. For example, the payment transaction tag is used to distinguish between different remote payment transactions.

Step S112, for the jth remote payment item in the recommended payment service terminal, obtaining a payment item statistics time period corresponding to the jth remote payment item. For example, the payment statistics period may be flexibly designed according to different service scenarios, such as several hours, several days, etc., without limitation.

Step S113, for the jth remote payment item in the recommended payment service terminal, determining the payment item statistical information corresponding to the jth remote payment item according to the payment item statistical time period corresponding to the jth remote payment item and the payment item tag corresponding to the jth remote payment item.

In this way, based on the contents described in the above steps S111 to S113, the payment item tag and the payment item statistic time period can be taken into account when determining the payment item statistic information, so that the acquisition of the payment item statistic information can be completely, accurately realized in real time.

Further, for step S120, the determining i payment security verification information according to the i payment transaction statistics information may include the following contents described in steps S121 to S123.

Step S121, determining, for a jth remote payment item in the recommended payment service terminal, payment item object information corresponding to the jth remote payment item according to the payment item statistical information of the jth remote payment item, where the jth remote payment item is any one of the i remote payment items, and j is an integer greater than 0 and less than or equal to i. For example, the payment item object information may be a participant or associated participant of the corresponding remote payment item, which may also be a payment terminal.

Step S122, for the jth remote payment item in the recommended payment service terminal, obtaining a payment item statistics time period corresponding to the jth remote payment item.

Step S123, for the jth remote payment item in the recommended payment service terminal, determining the payment security verification information corresponding to the jth remote payment item according to the payment item statistics time period corresponding to the jth remote payment item and the payment item object information corresponding to the jth remote payment item.

In this way, when the contents described in the above steps S121 to S123 are applied, the payment item statistic period and the payment item object information can be considered when determining the payment security verification information, and the related parties in the payment item object information can be further considered, so that the credibility of the payment security verification information can be improved by using the risk transfer concept.

In practical implementation, the determining i security verification tags according to the i payment security verification information described in step S130 may include the following steps S131 and S132.

Step S131, for a jth remote payment transaction in the recommended payment service terminal, obtaining a payment transaction verification indicator corresponding to the jth remote payment transaction, where the jth remote payment transaction is any one of the i remote payment transactions, and j is an integer greater than 0 and less than or equal to i. For example, the payment transaction verification index includes multiple levels of verification indexes, such as a verification index for a payment period, a verification index for a payment object, a verification index for a payment method, and the like.

Step S132, for the jth remote payment transaction in the recommended payment service terminal, determining a security verification tag corresponding to the jth remote payment transaction according to a payment transaction verification index corresponding to the jth remote payment transaction and payment security verification information corresponding to the jth remote payment transaction.

It can be understood that, when the contents described in the above steps S131 and S132 are implemented, the payment security verification information can be classified and analyzed through the payment item verification index, so as to distinguish different verification results of the payment security verification information, thereby implementing accurate screening and determination of the security verification index, and further improving the distinction degree between the security verification tags.

For a possible embodiment, the step of acquiring, in step S140, i payment information detection results corresponding to the recommended payment service terminal through a payment information detection model based on the i security verification tags may include the following steps S141 and S142.

Step S141, based on the i security verification tags, obtaining i remote payment verification records through at least one verification record determination unit included in the payment information detection model. For example, the check record determination unit may be one of network layers of the payment information detection model, and the remote payment check record is used for characterizing the check record corresponding to the security verification tag.

Step S142, based on the i remote payment verification records, obtaining i payment information detection results through at least one payment information detection unit included in the payment information detection model. For example, the payment information detection unit may be another type of network layer of the payment information detection model.

In this way, the remote payment verification record and the payment information detection result can be successively acquired based on different model units of the payment information detection model through the steps S141 and S142, so that the accuracy of the payment information detection result is not affected by mutual interference among model parameters in the process of acquiring the payment information detection result.

The inventor finds that the payment information detection model is used as a key for obtaining the payment information detection result, and the quality of training of the payment information detection model is directly related to the reliability of the payment information detection result, so that the payment information detection model needs to be trained sufficiently before being used to ensure the operation stability of the payment information detection model. To achieve this, before the step of obtaining i payment information detection results corresponding to the recommended payment service terminal through a payment information detection model based on the i security verification tags, which is described in step S140, the method may further include the contents described in step S210 to step S260.

Step S210, obtaining statistical information of the to-be-trained payment items corresponding to each to-be-trained payment item in a historical payment detection record to obtain i statistical information of the to-be-trained payment items, wherein the historical payment detection record comprises the i to-be-trained payment items, and the to-be-trained payment items and the statistical information of the to-be-trained payment items have a one-to-one correspondence relationship. For example, the historical payment detection records may be obtained from a database of a big data server, or may be obtained from other databases, which is not limited herein.

Step S220, i pieces of payment safety verification information to be trained are determined according to the i pieces of payment statistical information to be trained, wherein the payment safety verification information to be trained and the payment safety verification information to be trained have a one-to-one correspondence relationship.

Step S230, determining i to-be-trained safety verification tags according to the i to-be-trained payment safety verification information, where the to-be-trained safety verification tags and the to-be-trained payment items have a one-to-one correspondence relationship.

Step S240, based on the i to-be-trained security verification tags, obtaining i to-be-trained payment information detection results corresponding to the historical payment detection records through a to-be-trained payment information detection model, where the to-be-trained payment information detection results and the to-be-trained payment items have a one-to-one correspondence relationship.

Step S250, i proper payment information detection results corresponding to the proper detection model are obtained.

Step S260, training the payment information detection model to be trained according to the i regular example payment information detection results and the i payment information detection results to be trained until set model training conditions are met, and obtaining the payment information detection model.

By such design, based on the steps S210 to S260, the payment information detection model can be trained sufficiently to ensure the operation stability of the payment information detection model, so as to ensure the training quality of the payment information detection model, and further ensure the reliability of the payment information detection result obtained by using the payment information detection model.

Further, the step S250 of obtaining the detection results of the i proper example payment information corresponding to the proper example detection model may include steps S251 to S254.

Step S251, obtaining the statistical information of the sound payment items corresponding to each sound payment item in the sound detection model to obtain i pieces of statistical information of the sound payment items, where the sound detection model includes i pieces of sound payment items, and the sound payment items and the statistical information of the sound payment items have a one-to-one correspondence relationship.

Step S252, determining i pieces of proper payment security verification information according to the i pieces of proper payment statistics information, where the proper payment security verification information and the proper payment statistics information have a one-to-one correspondence relationship.

Step S253, determining i good-case security verification tags according to the i good-case payment security verification information, where the good-case security verification tags and the good-case payment items have a one-to-one correspondence relationship.

Step S254, based on the i proper example security verification tags, obtaining i proper example payment information detection results corresponding to the proper example detection model through the payment information detection model to be trained, where the proper example payment information detection results and the proper example payment items have a one-to-one correspondence relationship.

Further, the step S260 of training the to-be-trained payment information detection model according to the i positive example payment information detection results and the i to-be-trained payment information detection results until a set model training condition is met to obtain the payment information detection model may include the following steps S261 to S263.

Step S261 calculates, for each payment to be trained in the historical payment detection record and the corresponding regular payment item of each payment to be trained, a detection result difference value between the payment information detection result to be trained and the regular payment information detection result by using a payment security evaluation index, to obtain i detection result difference values. For example, the payment security evaluation index may be obtained according to risk behavior information of a payment security risk that has occurred before, which is not described herein again. The detection result difference value is used for representing the difference between different detection results, and the larger the detection result difference value is, the larger the difference between different detection results is.

Step S262, updating the model parameters of the payment information detection model to be trained according to the i detection result difference values. For example, the model parameters include conventional parameters of a neural network model, and are not described in detail herein.

And step S263, if the set model training condition is met, acquiring the payment information detection model according to the updated model parameters. And the set model training condition is that the difference value of each detection result is smaller than the set difference value. For example, the set difference value may be adjusted according to actual conditions, and is not limited herein.

By the design, iterative training can be performed on the payment information detection model to be trained according to the difference values of the detection results, so that the training quality of the payment information detection model is ensured.

In another possible embodiment, in order to ensure the comprehensiveness of model training, a positive example and a negative example may be further combined to perform collaborative training, and to achieve this, before the payment information detection model obtains i payment information detection results corresponding to the recommended payment service terminal based on the i security verification tags in step S140, the method may further include the following contents described in step S310 to step S360.

Step S310, obtaining statistical information of the to-be-trained payment items corresponding to each to-be-trained payment item in a historical payment detection record to obtain i statistical information of the to-be-trained payment items, wherein the historical payment detection record comprises the i to-be-trained payment items, and the to-be-trained payment items and the statistical information of the to-be-trained payment items have a one-to-one correspondence relationship.

Step S320, determining i pieces of payment security verification information to be trained according to the i pieces of payment statistics information to be trained, where the payment security verification information to be trained and the payment statistics information to be trained have a one-to-one correspondence relationship.

Step S330, i to-be-trained safety verification labels are determined according to the i to-be-trained payment safety verification information, wherein the to-be-trained safety verification labels and the to-be-trained payment items have one-to-one correspondence.

Step S340, based on the i to-be-trained security verification tags, obtaining i to-be-trained payment information detection results corresponding to the historical payment detection records through a to-be-trained payment information detection model, where the to-be-trained payment information detection results and the to-be-trained payment items have a one-to-one correspondence relationship.

Step S350, acquiring i regular example payment information detection results corresponding to the regular example detection model; and acquiring the detection results of the i negative example payment information corresponding to the negative example detection model.

Step S360, training the payment information detection model to be trained according to the i positive payment information detection results, the i negative payment information detection results and the i payment information detection results to be trained until set model training conditions are met, and obtaining the payment information detection model. The positive detection model is used for detecting payment information of the payment terminal without payment safety risk, and the negative detection model is used for detecting the payment information of the payment terminal with payment safety risk;

by the design, based on the steps S310 to S360, the payment information detection model can be trained by combining the positive detection model and the negative detection model, so that the comprehensiveness of model training is ensured, and the problem of detection omission in subsequent model use is avoided.

Further, in step S350, the obtaining i proper payment information detection results corresponding to the proper detection model includes: acquiring the statistical information of the payment matters of the positive example corresponding to each payment matter of the positive example in a positive example detection model to obtain the statistical information of i payment matters of the positive example, wherein the positive example detection model comprises i payment matters of the positive example, and the payment matters of the positive example and the statistical information of the payment matters of the positive example have a one-to-one correspondence relationship; determining i pieces of positive example payment safety verification information according to the i pieces of positive example payment statistical information, wherein the positive example payment safety verification information and the positive example payment statistical information have a one-to-one correspondence relationship; determining i good example security verification tags according to the i good example payment security verification information, wherein the good example security verification tags have a one-to-one correspondence relationship with the good example payment matters; based on the i good example security verification tags, i good example payment information detection results corresponding to the good example detection model are obtained through the payment information detection model to be trained, wherein the good example payment information detection results and the good example payment items have a one-to-one correspondence relationship.

Further, in step S350, the obtaining i negative example payment information detection results corresponding to the negative example detection model includes: acquiring negative example payment statistical information corresponding to each negative example payment in a negative example detection model to obtain i negative example payment statistical information, wherein the negative example detection model comprises i negative example payments, and the negative example payments and the negative example payment statistical information have a one-to-one correspondence relationship; determining i negative example payment safety verification information according to the i negative example payment statistical information, wherein the negative example payment safety verification information and the negative example payment statistical information have a one-to-one correspondence relationship; determining i negative example security verification tags according to the i negative example payment security verification information, wherein the negative example security verification tags have a one-to-one correspondence relationship with the negative example payment matters; and acquiring i negative example payment information detection results corresponding to the negative example detection model through the to-be-trained payment information detection model based on the i negative example security verification labels, wherein the negative example payment information detection results and the negative example payment items have a one-to-one correspondence relationship.

Further, the step S360 of training the to-be-trained payment information detection model according to the i positive payment information detection results, the i negative payment information detection results, and the i to-be-trained payment information detection results until a set model training condition is met to obtain the payment information detection model may include the following steps S361 to S363.

Step S361, for each payment to be trained in the historical payment detection record and the corresponding regular payment item of each payment to be trained, calculating a first detection result difference value between the detection result of the payment to be trained and the detection result of the regular payment information by using a first payment security evaluation index, and obtaining i first detection result difference values.

Step S362, for each payment to be trained and the negative payment in the historical payment detection record, calculating a second detection result difference value between the payment information detection result to be trained and the negative payment information detection result by using a second payment safety evaluation index, and obtaining i second detection result difference values.

Step S363, updating the model parameters of the payment information detection model to be trained according to the i first detection result difference values and the i second detection result difference values; if the set model training condition is met, acquiring the payment information detection model according to the updated model parameters; the set model training condition is that the mean value of the i first detection result difference values and the i second detection result difference values is smaller than a set difference value.

It is to be understood that, in the above embodiments, the first training mode using only positive examples for training and the second training mode using positive examples for training can be used as one mode, and are not limited herein.

In a real-time process, the inventor finds that the detection of the payment security risk is related to the payment operation data, and therefore, in order to ensure the detection accuracy of the payment security risk, the payment operation data of the recommended payment service terminal needs to be considered, and for this purpose, the determination of whether the recommended payment service terminal has the payment security risk according to the i payment information detection results described in step S150 may include the following steps S151 to S157.

Step S151, obtaining a payment operation data set of the recommended payment service terminal corresponding to the i payment information detection results, where the payment operation data set includes at least one set of payment operation data and operation behavior mark information corresponding thereto, and the operation behavior mark information includes an operation behavior corresponding to the payment operation data and real-time mark information corresponding thereto describing the operation behavior. For example, the payment operation data may be touch operation data or voice operation data of the user, and is not limited herein.

Step S152, inputting the payment operation data set into a preset data feature extraction thread, so that the preset data feature extraction thread outputs a feature extraction result of payment operation data according to the payment operation data, where the feature extraction result of the payment operation data includes an operation behavior corresponding to the payment operation data extracted by the preset data feature extraction thread and operation behavior feature information of the operation behavior. For example, the data feature extraction thread may be configured according to actual requirements, which is not described herein. The operational behavior characteristic information may be used to distinguish between different operational behaviors.

Step S153, determining first overlapping feature information of the real-time tag information in the operation behavior tag information and the operation behavior feature information in the feature extraction result of the payment operation data. For example, the first overlay characteristic information may be information that a cross or an association exists between the real-time marking information and the operation behavior characteristic information.

Step S154, constructing first thread training information based on the feature extraction result of the payment operation data, the operation behavior flag information, and the first overlapping feature information, which are output by the preset data feature extraction thread currently. For example, the first thread training information is used to train a data feature extraction thread.

Step S155, training the preset data feature extraction thread according to the first thread training information, and obtaining the trained preset data feature extraction thread when the training frequency satisfies a set training frequency. For example, the set training times are adjusted according to actual business requirements, and are not limited herein.

Step S156 is to input the request data corresponding to the remote payment initiation request corresponding to the recommended payment service terminal into the trained preset data feature extraction thread, and output the feature extraction result of the request data corresponding to the remote payment initiation request corresponding to the recommended payment service terminal. For example, the remote payment initiation request may be a passive request or an active request, and is not limited herein.

Step S157, determining whether the recommended payment service terminal is in a data interface verification-free state in the current time period according to the feature extraction result of the payment operation data and the feature extraction result of the request data; if yes, judging that the recommended payment service terminal has a payment safety risk; if not, the recommended payment service terminal is judged to have no payment safety risk. For example, the current time period may be adjusted according to the payment frequency of the recommended payment service terminal, where the higher the payment frequency is, the shorter the current time period is, the verification-free state of the data interface is used to represent that the data interface of the recommended payment service terminal is in an open state, which may cause the data information of the recommended payment service terminal to be illegally accessed.

In this way, based on the contents described in the above steps S151 to S157, the payment operation data of the recommended payment service terminal can be taken into account, so as to implement the detection of the payment security risk based on the data interface layer, thereby not only reducing the processing resources required for the payment key verification, but also ensuring the detection accuracy and timeliness of the payment security risk, and avoiding the data information of the recommended payment service terminal from being illegally accessed due to the detection delay.

In an alternative embodiment, when it is determined that the recommended payment service terminal has the payment security risk, the method may further include: sending prompt information to the recommended payment service terminal to prompt the recommended payment service terminal to adjust the verification-free state of the corresponding data interface; and the corresponding data interface is the data interface corresponding to the encryption information of the recommended payment service terminal. Thus, the data information of the recommended payment service terminal can be prevented from being illegally accessed.

In an alternative embodiment, the determining the first overlapping feature information of the real-time tag information in the operation behavior tag information and the operation behavior feature information in the feature extraction result of the payment operation data, which is described in step S153, includes: acquiring a fusion information set of the real-time marking information and the operation behavior characteristic information, and selecting at least three groups of first fusion information from the fusion information set; acquiring first fusion weight according to the first fusion information; acquiring second fusion information corresponding to a second fusion weight with the largest weight difference value with the first fusion weight in the fusion information set; if the second fusion information is in the fusion information subset matched with the first fusion weight, determining the first overlapping feature information according to the first fusion weight; and if the second fusion information is not in the fusion information subset matched with the first fusion weight, selecting at least three groups of fusion information from the first fusion information and the second fusion information as the first fusion information and acquiring the first fusion weight again.

In an alternative embodiment, the constructing, by the step S154, the first thread training information based on the feature extraction result of the payment operation data output by the preset data feature extraction thread, the operation behavior marking information, and the first overlapping feature information includes: acquiring time sequence feature superposition information and time sequence feature difference information of the real-time marking information and the operation behavior feature information, and acquiring overlapping description information of the first overlapping feature information; and acquiring the first thread training information according to the time sequence feature coincidence information, the time sequence feature difference information and the overlapping description information of the first overlapping feature information.

In an alternative embodiment, the feature extraction of the operation behavior by the preset data feature extraction thread is based on a convolutional neural network, including at least one convolutional layer.

In an alternative embodiment, the training of the preset data feature extraction thread described in step S154 includes: acquiring a global operation data cleaning result and a local operation data cleaning result of the payment operation data after the payment operation data is cleaned through the preset data feature extraction thread data; acquiring a first cleaned data set with a first recognition degree as a characteristic recognition degree, a second cleaned data set with a second recognition degree as a characteristic recognition degree and a third cleaned data set with a third recognition degree as a characteristic recognition degree according to the overall operation data cleaning result; wherein the first, second, and third degrees of identification are different from each other, and the first, second, and third cleaned data sets are different from each other; acquiring a fourth cleaned data set with the characteristic identification degree of the first identification degree, a fifth cleaned data set with the characteristic identification degree of the second identification degree and a sixth cleaned data set with the characteristic identification degree of the third identification degree according to the local operation data cleaning result; wherein the fourth cleaned data set, the fifth cleaned data set, and the sixth cleaned data set are different from each other; splicing the first cleaned data set with the fourth cleaned data set to obtain a first spliced data set, splicing the second cleaned data set with the fifth cleaned data set to obtain a second spliced data set, and splicing the third cleaned data set with the sixth cleaned data set to obtain a third spliced data set; wherein the first, second, and third stitched data sets are different from each other; and acquiring a feature extraction result of the payment operation data according to the first splicing data set, the second splicing data set and the third splicing data set.

In an alternative embodiment, the acquiring of the payment operation data set of the recommended payment service terminal corresponding to the i payment information detection results described in step S151 includes: acquiring at least one first information detection index corresponding to each payment information detection result, and performing detection index attribute analysis on the first information detection index to obtain first mark information, wherein the first mark information comprises a first operation behavior corresponding to the first information detection index and first real-time mark information corresponding to the first operation behavior; performing operation behavior recognition on the first operation behavior, and then constructing a second information detection index; identifying and obtaining second mark information of the second information detection index according to the operation behavior; and acquiring the payment operation data set according to the first information detection index and the first mark information thereof, the second information detection index and the second mark information thereof.

Based on the same inventive concept, a payment mode recommendation system based on big data and a block chain is further provided, and the following is further described.

A payment mode recommendation system based on big data and a block chain comprises a cloud computing server and a payment service terminal which are communicated with each other; wherein the cloud computing server is to:

acquiring payment data transmission information and data interface updating information of a payment service terminal during online payment;

respectively training a data network analysis model and a data safety protection model by using the payment data transmission information and the data interface updating information, and correspondingly storing the data network analysis model, the data safety protection model and payment recommendation permission information of the payment service terminal into a preset database;

when the payment mode recommendation is performed on any payment service terminal, acquiring payment data transmission information and data interface updating information of the recommended payment service terminal, finding a data network analysis model matched with the acquired payment data transmission information from a preset database, acquiring first payment recommendation permission information corresponding to the found data network analysis model from the preset database, finding a data security protection model matched with the acquired data interface updating information from the preset database, and acquiring second payment recommendation permission information corresponding to the found data security protection model from the preset database;

comparing whether the first payment recommendation permission information and the second payment recommendation permission information are the same or not, and verifying whether the recommended payment service terminal passes recommendation verification or not according to a comparison result;

and when the recommended payment service terminal passes the recommendation verification, determining a recommended payment mode of the recommended payment service terminal according to the found data network analysis model and the found data security protection model.

It will be appreciated that reference may be made to the description of method embodiments in relation to the above description of system embodiments.

It should be understood that, for technical terms that are not noun-explained in the above, a person skilled in the art can deduce and unambiguously determine the meaning of the present invention from the above disclosure, for example, for some values, coefficients, weights, indexes, factors and other terms, a person skilled in the art can deduce and determine from the logical relationship between the above and the below, and the value range of these values can be selected according to the actual situation, for example, 0 to 1, for example, 1 to 10, and for example, 50 to 100, which is not limited herein.

The skilled person can unambiguously determine some preset, reference, predetermined, set and target technical features/terms, such as threshold values, threshold intervals, threshold ranges, etc., from the above disclosure. For some technical characteristic terms which are not explained, the technical solution can be clearly and completely implemented by those skilled in the art by reasonably and unambiguously deriving the technical solution based on the logical relations in the previous and following paragraphs. Prefixes of unexplained technical feature terms, such as "first", "second", "previous", "next", "current", "history", "latest", "best", "target", "specified", and "real-time", etc., can be unambiguously derived and determined from the context. Suffixes of technical feature terms not to be explained, such as "list", "feature", "sequence", "set", "matrix", "unit", "element", "track", and "list", etc., can also be derived and determined unambiguously from the foregoing and the following.

The foregoing disclosure of embodiments of the present invention will be apparent to those skilled in the art. It should be understood that the process of deriving and analyzing technical terms, which are not explained, by those skilled in the art based on the above disclosure is based on the contents described in the present application, and thus the above contents are not an inventive judgment of the overall scheme.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered merely illustrative and not restrictive of the broad application. Various modifications, improvements and adaptations to the present application may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific terminology to describe embodiments of the application. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the present application is included in at least one embodiment of the present application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of at least one embodiment of the present application may be combined as appropriate.

In addition, those skilled in the art will recognize that the various aspects of the application may be illustrated and described in terms of several patentable species or contexts, including any new and useful combination of procedures, machines, articles, or materials, or any new and useful modifications thereof. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as a "unit", "component", or "system". Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in at least one computer readable medium.

A computer readable signal medium may comprise a propagated data signal with computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, electrical cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the execution of aspects of the present application may be written in any combination of one or more programming languages, including object oriented programming, such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, or similar conventional programming languages, such as the "C" programming language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages, such as Python, Ruby, and Groovy, or other programming languages. The programming code may execute entirely on the user's computer, as a stand-alone software package, partly on the user's computer, partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order of the process elements and sequences described herein, the use of numerical letters, or other designations are not intended to limit the order of the processes and methods unless otherwise indicated in the claims. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it should be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware means, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

It should also be appreciated that in the foregoing description of embodiments of the present application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the invention. However, this method of disclosure is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Claims

1. A payment mode recommendation method based on big data and a block chain is characterized by comprising the following steps:

2. The method of claim 1,

wherein the verifying whether the recommended payment service terminal passes the recommendation verification according to the comparison result comprises: when the comparison result is that the first payment recommendation permission information is the same as the second payment recommendation permission information, determining that the recommended payment service terminal passes recommendation verification; when the comparison result is that the first payment recommendation permission information is different from the second payment recommendation permission information, determining that the recommended payment service terminal does not pass recommendation verification;

the acquiring payment data transmission information and data interface update information of the payment service terminal when performing online payment further comprises: acquiring service processing information uploaded by a payment service terminal during online payment;

wherein the correspondingly storing the data network analysis model, the data security protection model and the payment recommendation permission information of the payment service terminal into a preset database comprises: and correspondingly storing a data network analysis model, a data security protection model, the service processing information and payment recommendation permission information of the payment service terminal into a preset database.

3. The method of claim 2, wherein the verifying whether the recommended payment service terminal passes the recommendation verification according to the comparison result comprises:

when the comparison result is that the first payment recommendation permission information is the same as the second payment recommendation permission information, acquiring service processing information corresponding to the first payment recommendation permission information from the preset database, and verifying whether the recommended payment service terminal passes recommendation verification or not by using the acquired service processing information;

and when the comparison result shows that the first payment recommendation permission information is different from the second payment recommendation permission information, determining that the recommended payment service terminal does not pass the recommendation verification, or respectively acquiring service processing information corresponding to the first payment recommendation permission information and the second payment recommendation permission information, and verifying whether the recommended payment service terminal passes the recommendation verification or not by using the service processing information corresponding to the first payment recommendation permission information and the service processing information corresponding to the second payment recommendation permission information.

4. The method of claim 3,

the verifying whether the recommended payment service terminal passes the recommendation verification by using the acquired service processing information includes: performing interface update state analysis on the acquired data interface update information of the recommended payment service terminal, comparing whether the result obtained by the interface update state analysis is consistent with the acquired service processing information, if so, determining that the recommended payment service terminal passes recommendation verification, otherwise, determining that the recommended payment service terminal does not pass recommendation verification;

or

The verifying whether the recommended payment service terminal passes the recommendation verification by using the acquired service processing information includes: calculating the matching degree of the payment service between the collected data interface updating information of the recommended payment service terminal and the obtained service processing information; and identifying whether the corresponding matching degree calculation result is greater than a set threshold value, if so, determining that the recommended payment service terminal passes the recommendation verification, otherwise, determining that the recommended payment service terminal does not pass the recommendation verification.

5. The method of claim 3, wherein the verifying whether the recommended payment service terminal passes the recommendation verification by using the service processing information corresponding to the first payment recommendation permission information and the service processing information corresponding to the second payment recommendation permission information comprises:

calculating the matching degree of the payment service of the collected data interface updating information of the recommended payment service terminal and the service processing information corresponding to the first payment recommendation permission information to obtain the matching degree of the first payment service;

calculating the matching degree of the payment service for the collected data interface updating information of the recommended payment service terminal and the service processing information corresponding to the second payment recommendation permission information to obtain a second payment service matching degree;

if the matching degree of the first payment service is greater than the matching degree of the second payment service and the matching degree of the first payment service is also greater than the set threshold value, determining that the recommended payment service terminal is the payment service terminal corresponding to the first payment recommendation permission information and determining that the payment service terminal corresponding to the first payment recommendation permission information passes the recommendation verification, otherwise, determining that the payment service terminal corresponding to the first payment recommendation permission information does not pass the recommendation verification;

and if the second payment service matching degree is greater than the first payment service matching degree and the second payment service matching degree is also greater than the set threshold, determining that the recommended payment service terminal is the payment service terminal corresponding to the second payment recommendation permission information, and determining that the payment service terminal corresponding to the second payment recommendation permission information passes the recommendation verification, otherwise, determining that the payment service terminal corresponding to the second payment recommendation permission information does not pass the recommendation verification.

6. The method of claim 1, wherein determining the recommended payment mode of the recommended payment service terminal according to the found data network analysis model and the found data security model comprises:

acquiring current payment environment data of the recommended payment service terminal according to the found data network analysis model and the found data security protection model;

obtaining effective behavior data of payment business behavior in the current payment environment data;

extracting behavior data characteristics of the effective behavior data;

according to a password payment classification model, a fingerprint payment classification model and a face-brushing payment classification model, classifying password payment, fingerprint payment and face-brushing payment of behavior data features of the effective behavior data to obtain password payment behavior features, fingerprint payment behavior features and face-brushing payment behavior features of the payment business behaviors;

and determining the recommended payment mode of the recommended payment service terminal according to the password payment behavior characteristic, the fingerprint payment behavior characteristic and the face brushing payment behavior characteristic.

7. The method of claim 6,

before obtaining the valid behavior data of the payment service behavior in the current payment environment data, the method includes: sequentially establishing a password payment classification model, a fingerprint payment classification model and a face brushing payment classification model;

wherein, establish password payment classification model, fingerprint payment classification model and brush face payment classification model in proper order, include:

acquiring a training sample set;

obtaining effective behavior data of each current payment environment data in the training sample set;

extracting behavior data characteristics of each effective behavior data;

according to the behavior data characteristics of the effective behavior data of all current payment environment data in the training sample set, sequentially establishing a password payment classification model, a fingerprint payment classification model and a face brushing payment classification model;

further, the establishing a password payment classification model according to the behavior data characteristics in the training sample set includes:

dividing the training sample set into pure letter password payment training samples, pure digital password payment training samples and mixed password payment training samples according to three types of sample tags of the pure letter password payment sample tags, the pure digital password payment sample tags and the mixed password payment sample tags;

respectively training the behavior data characteristics of the pure letter password payment training sample, the behavior data characteristics of the pure digital password payment training sample and the behavior data characteristics of the mixed password payment training sample to obtain three classification models corresponding to the password payment, wherein the three classification models respectively correspond to a pure letter password payment sample label, a pure digital password payment sample label and a mixed password payment sample label;

further, the establishing a fingerprint payment classification model according to the behavior data characteristics of the training sample set includes:

dividing the pure letter password payment training sample, the pure digital password payment training sample and the mixed password payment training sample again according to fingerprint payment statistical information to obtain the fingerprint payment sample;

respectively training the behavior data characteristics of the fingerprint payment samples to obtain two classification models corresponding to the fingerprint payment samples, wherein the two classification models respectively correspond to point contact payment and press type payment;

further, according to the behavior data characteristics of the training sample set, a face brushing payment classification model is established, and the method comprises the following steps:

respectively carrying out secondary division on the associated behavior data characteristics corresponding to the fingerprint payment samples according to a face brushing payment division index to obtain a left half face region training sample, a right half face region training sample, an upper half face region training sample and a lower half face region training sample;

training the left half face region training sample, the right half face region training sample, the upper half face region training sample and the lower half face region training sample to construct a first-stage face brushing payment classification model;

training the behavior data characteristics of the target facial region training sample corresponding to the set facial feature recognition degree by taking the set facial feature recognition degree as a reference grade under the divided facial region training samples to obtain a second-stage face brushing payment classification model;

training a third-level face brushing payment classification model with a dynamic face region fitting function aiming at each target face region training sample with set face feature recognition degree under the divided face region training samples;

the obtaining of the effective behavior data of the payment service behavior in the current payment environment data includes:

detecting payment business behaviors in the current payment environment data, and determining execution parameters of a behavior function of the payment business behaviors;

correcting the original behavior data of the payment business behavior based on the execution parameters of the behavior function of the payment business behavior, and analyzing and setting the payment business behavior in a payment time interval by taking the behavior function of the payment business behavior as a reference to obtain effective behavior data of the payment business behavior;

wherein the extracting the behavior data feature of the valid behavior data comprises: and extracting the multidimensional description characteristics of the behavior data of the payment business behavior aiming at the effective behavior data of the payment business behavior.

8. The method according to claim 6, wherein the classifying the behavior data features of the valid behavior data according to a password payment classification model, a fingerprint payment classification model and a face-brushing payment classification model in turn to obtain the password payment behavior features, the fingerprint payment behavior features and the face-brushing payment behavior features of the payment service behaviors comprises:

performing password payment classification on the behavior data characteristics of the effective behavior data according to a password payment classification model to obtain password payment behavior characteristics of the payment business behavior;

selecting a fingerprint payment classification model corresponding to the password payment behavior characteristics of the payment business behavior, and performing fingerprint payment classification on the behavior data characteristics of the effective behavior data according to the fingerprint payment classification model to obtain the fingerprint payment behavior characteristics of the payment business behavior;

selecting a face-brushing payment classification model corresponding to the password payment behavior characteristics and the fingerprint payment behavior characteristics of the payment business behavior, and performing face-brushing payment classification on the behavior data characteristics of the effective behavior data according to the face-brushing payment classification model to obtain face-brushing payment behavior characteristics of the payment business behavior;

selecting a face-brushing payment classification model corresponding to password payment behavior characteristics and fingerprint payment behavior characteristics of the payment business behaviors, and performing face-brushing payment classification on behavior data characteristics of effective behavior data according to the face-brushing payment classification model to obtain face-brushing payment behavior characteristics of the payment business behaviors, wherein the face-brushing payment classification model comprises the following steps:

selecting a first-level face-brushing payment classification model corresponding to password payment behavior characteristics and fingerprint payment behavior characteristics of the payment business behavior, and inputting behavior data characteristics of the effective behavior data into the first-level face-brushing payment classification model to obtain first-level face area dynamic characteristics corresponding to the effective behavior data and face behavior characteristic weights corresponding to the first-level face area dynamic characteristics;

selecting a second-level face brushing payment classification model according to the first-level face region dynamic feature with the maximum corresponding face behavior feature weight corresponding to the first-level face region dynamic feature, and inputting the behavior data feature of the effective behavior data into the second-level face brushing payment classification model to obtain the second-level face region dynamic feature corresponding to the effective behavior data and the face behavior feature weight corresponding to each second-level face region dynamic feature;

selecting a third-level face brushing payment classification model according to the second-level face region dynamic characteristics with the maximum facial behavior characteristic weight corresponding to each second-level face region dynamic characteristic, and inputting the behavior data characteristics of the effective behavior data into the third-level face brushing payment classification model to obtain third-level face region dynamic characteristics corresponding to the effective behavior data;

and obtaining the face brushing payment behavior characteristics of the payment business behavior according to the third-level face region dynamic characteristics, the face behavior characteristic weight corresponding to the first-level face region dynamic characteristics and the face behavior characteristic weight corresponding to the second-level face region dynamic characteristics.

9. The method according to claim 1, wherein determining the recommended payment mode of the recommended payment service terminal through the password payment behavior characteristic, the fingerprint payment behavior characteristic and the face-brushing payment behavior characteristic comprises:

acquiring a first payment behavior fusion feature set according to the password payment behavior feature, the fingerprint payment behavior feature and the face brushing payment behavior feature, wherein the first payment behavior fusion feature set comprises n first payment behavior fusion features, each first payment behavior fusion feature has m payment behavior feature distribution information, n is an integer greater than 1, and m is an integer greater than 1;

generating a first recommended payment behavior feature set according to the first payment behavior fusion feature set, wherein the first recommended payment behavior feature set comprises n first recommended payment behavior features, each first recommended payment behavior feature is obtained after feature screening is carried out on the first payment behavior fusion feature, and each first recommended payment behavior feature has m pieces of recommended feature distribution information;

determining a first payment mode recommendation index according to the first recommended payment behavior feature set aiming at target payment behavior feature distribution information, wherein the target payment behavior feature distribution information belongs to any one recommended feature distribution information in the m recommended feature distribution information;

regarding the target payment behavior characteristic distribution information, if a payment safety verification condition is met, taking the first payment mode recommendation index as a current recommendation index of the recommended payment service terminal corresponding to the target payment behavior characteristic distribution information;

determining a recommended payment mode of the recommended payment service terminal according to the current recommendation index;

the method for determining the recommended payment mode of the recommended payment service terminal through the current recommendation index comprises the following steps:

and determining the recommendation index of each candidate payment mode according to the current recommendation index, and determining the candidate payment mode with the maximum recommendation index as the recommended payment mode, wherein the candidate payment modes comprise a password payment mode, a fingerprint payment mode and a face brushing payment mode.

10. A cloud computing server comprising a processing engine, a network module, and a memory; the processing engine and the memory communicate through the network module, the processing engine reading a computer program from the memory and operating to perform the method of any of claims 1-9.