Risk transaction processing method and device, equipment and storage medium

Abstract

The invention discloses a risk transaction processing method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: inputting the risk data of a current risk transaction into a pre-trained risk classification model, so as to enable the risk classification model to determine the risk channel of the current risk transaction; determining a risk model according to the risk data of the current risk transaction; and inputting the risk channel and the risk model of the current risk transaction into a pre-trained risk processing model, so that the risk processing model determines a processing mode of the current risk transaction. According to the technical scheme, after the risk channel and the risk model of the current risk transaction are determined, the risk channel and the risk model of the current risk transaction are input into the risk processing model, and the processing mode of the current risk transaction is determined based on the risk processing model, so that the processing mode of the risk transaction is rapidly and accurately determined.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of artificial intelligence technology, and in particular, to a method, device, equipment and storage medium for processing risky transactions. Background Art

[0002] When managing and controlling risky transactions in the financial sector, risky transactions generated by different risk channels need to be handled differently. The risk channels of risky transactions are often difficult to determine. After identifying risky transactions in this scenario, risk handlers find it difficult to determine whether the risky transactions are caused by customers, customer bank cards, or merchants, making it difficult to use the correct handling method to handle risky transactions. Therefore, determining the risk channels and handling methods for risky transactions has become a technical problem that needs to be solved urgently.

[0003] In the existing technology, risk channels of risky transactions are usually identified manually, and the processing method is determined according to the risk channels. This has the disadvantages of low efficiency, high requirements on personnel, and low accuracy. Summary of the invention

[0004] The present invention provides a method, device, equipment and storage medium for processing risky transactions, so as to realize accurate and rapid risky transaction processing and reduce transaction risks.

[0005] In a first aspect, an embodiment of the present invention provides a method for processing risky transactions, including:

[0006] Inputting the risk data of the current risk transaction into a pre-trained risk classification model so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions;

[0007] Determining a risk model based on the risk data of the current risk transaction;

[0008] The risk channel and risk model of the current risk transaction are input into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

[0009] The technical solution of an embodiment of the present invention provides a method for processing risk transactions, including: inputting risk data of a current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on risk data and risk channels of historical risk transactions and risk data of real-time risk transactions; determining a risk model based on the risk data of the current risk transaction; inputting the risk channel and risk model of the current risk transaction into a pre-trained risk processing model, so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions. The above technical scheme, after acquiring the risk transaction data of the current risk transaction, can input the risk data of the current risk transaction into the risk classification model obtained by training the semi-supervised generative adversarial network with the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions, determine the risk channel of the current risk transaction based on the risk classification model, and realize rapid and accurate determination of the risk channel of the risk transaction, and can also determine the risk model of the current risk transaction based on the risk data of the current risk transaction, and then input the risk channel and risk model of the current risk transaction into the risk processing model obtained by training the decision tree with the risk model, risk channel and processing method of historical risk transactions, and determine the processing method of the current risk transaction based on the risk processing model, and realize rapid and accurate determination of the processing method of the risk transaction.

[0010] Furthermore, the risk channel is card risk, customer risk, merchant risk or equipment risk.

[0011] Furthermore, the training process of the risk classification model includes:

[0012] Acquire the risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions;

[0013] The risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions are used as training data to train the semi-supervised generative adversarial network until a first preset end condition is met, thereby obtaining the risk classification model.

[0014] Furthermore, the risk model is at least one of a long-term inactive account model, a money laundering model, a low credit model, a one-person-multiple-account model, a marketing model and a black device model.

[0015] Further, determining a risk model according to the risk data of the current risky transaction includes:

[0016] Determining whether the current risk transaction satisfies the risk conditions of each risk model according to the risk data of the current risk transaction;

[0017] In the case where it is determined that the current risk transaction satisfies the risk condition of the risk model, the risk model is determined as the risk model of the current risk transaction.

[0018] Determining whether the current risk transaction satisfies the risk conditions of each risk model according to the risk data of the current risk transaction;

[0019] In the case where it is determined that the current risk transaction satisfies the risk condition of the risk model, the risk model is determined as the risk model of the current risk transaction.

[0020] Furthermore, the training process of the risk management model includes:

[0021] Obtain risk models, risk channels, and processing methods of historical risk transactions, use the risk models and risk channels of each of the historical risk transactions as risk features, and use the processing methods of each of the historical risk transactions as label information to construct a training set;

[0022] Determine the target feature according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, and determine the sample set corresponding to the left node and the sample set corresponding to the right node of the current node according to the target feature and the feature value corresponding to the target feature, wherein the sample set corresponding to the root node of the decision tree is the training set, and the sample sets corresponding to the internal nodes are all subsets of the training set;

[0023] The left node and the right node are respectively used as the current node, and the target feature is determined according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, until the second preset end condition is met, so as to obtain the risk processing model.

[0024] In a second aspect, an embodiment of the present invention further provides a risk transaction processing device, including:

[0025] A classification module, used for inputting the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions;

[0026] A determination module, configured to determine a risk model according to the risk data of the current risky transaction;

[0027] A processing module is used to input the risk channel and risk model of the current risk transaction into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

[0028] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0029] at least one processor; and a memory communicatively coupled to the at least one processor;

[0030] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the risk transaction processing method as described in any one of the first aspects.

[0031] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute a method for processing risk transactions as described in any one of the first aspects.

[0032] In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the computer instructions are executed on a computer, the computer executes the method for processing risk transactions provided in the first aspect.

[0033] It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer-readable storage medium may be packaged together with the processor of the risk transaction processing device, or may be packaged separately from the processor of the risk transaction processing device, and this application does not limit this.

[0034] The description of the second, third, fourth and fifth aspects in this application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third, fourth and fifth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0035] In this application, the name of the risk transaction processing device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0036] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A flowchart of a method for processing risky transactions provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of the structure of a risk transaction processing device provided by an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0042] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0043] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0044] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0045] It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe various operations (or steps) as sequential processes, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.

[0046] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0047] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0048] Figure 1 The flowchart of a method for processing risky transactions provided by an embodiment of the present invention is applicable to situations where risky transactions need to be processed accurately and quickly. The method can be executed by a risky transaction processing device, such as Figure 1 As shown, the specific steps include:

[0049] Step 110: input the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction.

[0050] Among them, the risk data of the current risk transaction can be understood as the detailed data of the current risk transaction. The risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions. Historical risk transactions can be understood as risk transactions that occurred in the historical time period. The risk data of historical risk transactions are the transaction details of historical risk transactions. The risk channels of historical risk transactions are determined by risk processing personnel. Real-time risk transactions can be understood as risk transactions that are currently occurring. The semi-supervised generative adversarial network here can be an auxiliary classification generative adversarial network (ACGAN). The risk channel is card risk, customer risk, merchant risk or equipment risk.

[0051] In one embodiment, the training process of the risk classification model includes:

[0052] Acquire risk data and risk channels of historical risk transactions and risk data of the real-time risk transactions; use the risk data and risk channels of the historical risk transactions and risk data of the real-time risk transactions as training data to train the semi-supervised generative adversarial network until a first preset end condition is met, thereby obtaining the risk classification model.

[0053] The preset convergence condition may be at least one of satisfying the convergence of the loss function, the number of iterations being greater than a preset value, and the iteration time being greater than a preset time.

[0054] Specifically, first, the historical risk transactions that occurred in the historical time period and the risk channels of each historical risk transaction can be obtained from the risk monitoring system. The risk channels of each historical risk transaction are determined by the risk handling personnel and input into the risk monitoring system. The real-time risk transactions that are currently occurring can also be obtained from the risk transaction system. Then, the risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions can be input into ACGAN as training data, and semi-supervised training can be performed on ACGAN until the first preset end condition is met, and a risk classification model can be obtained.

[0055] The risk data of the current risk transaction is obtained from the risk monitoring system, and the risk data of the current risk transaction is input into the risk classification model obtained through the aforementioned training. The risk classification model can determine the risk channel of the current risk transaction by performing feature fitting on the risk data of the current risk transaction, thereby realizing rapid and accurate determination of the risk channel of the risk transaction.

[0056] In an embodiment of the present invention, the risk data of the current risk transaction is input into a risk classification model obtained by training a semi-supervised generative adversarial network with risk data and risk channels of historical risk transactions and risk data of real-time risk transactions, so that the risk classification model determines the risk channel of the current risk transaction, thereby achieving rapid and accurate determination of the risk channel of the risk transaction, and solving the defects of traditional risk channel determination methods, such as high personnel requirements, low timeliness, and low accuracy.

[0057] Step 120: Determine a risk model according to the risk data of the current risky transaction.

[0058] Among them, the risk model is at least one of a long-term inactive account model, a money laundering model, a low credit model, a one-person-multiple-account model, a marketing model and a black device model.

[0059] In one implementation, step 120 may specifically include:

[0060] Determine whether the current risk transaction satisfies the risk conditions of each of the risk models according to the risk data of the current risk transaction; and determine the risk model as the risk model of the current risk transaction if it is determined that the current risk transaction satisfies the risk conditions of the risk model.

[0061] Specifically, the risk model of the current risk transaction can be determined by comparing the risk data of the current risk transaction with the risk conditions of each risk model, and the current risk transaction can correspond to at least one risk model. Specifically, the risk data of the current risk transaction and the risk conditions of each risk model can be compared in sequence, and the risk model of the current risk transaction can be determined according to the comparison result.

[0062] The risk condition of the long-term inactive account model is: the trading account has no transaction record in the historical time period of the preset length or the transaction fluctuation range is less than the preset range, where the preset length and the preset range can be set according to actual needs and are not specifically limited here; the risk condition of the money laundering model is: at least one of the transaction flow, transaction amount and transaction counterparty is abnormal, where the abnormal conditions of the transaction flow, transaction amount and transaction counterparty can be set according to actual needs and are not specifically limited here; the risk condition of the low credit model is: the trading account is a low credit account, where the judgment conditions of the low credit account can be set according to actual needs , no specific limitation is made here; the risk condition of the one-person-multiple-account model is: the user corresponding to the transaction account has more than a first preset number of transaction accounts, wherein the specific value of the first preset number can be set according to actual needs, and no specific limitation is made here; the risk condition of the marketing model is: the counterparty is a merchant and the merchant has more than a second preset number of risky transactions, wherein the specific value of the second preset number can be set according to actual needs, and no specific limitation is made here; the risk condition of the black device model is that the transaction device is a risky device, wherein the judgment condition of the risky device can be set according to actual needs, and no specific limitation is made here.

[0063] If the comparison result between the risk data of the current risk transaction and the risk conditions of the risk model is consistent, the risk model is determined as the risk model corresponding to the current risk transaction, that is, if the transaction account corresponding to the current risk transaction has no transaction record within the historical time period of a preset length or the transaction fluctuation range is less than the preset range, the risk model is determined to be a long-term immobile account model; if at least one of the transaction flow, transaction amount and transaction counterparty of the current risk transaction is abnormal, the risk model is determined to be a money laundering model; if the transaction account of the current risk transaction is a low-credit account, the risk model is determined to be a low-credit model; if the user corresponding to the transaction account of the current risk transaction has more than a first preset number of transaction accounts, the risk model is determined to be a one-person multiple-account model; if the counterparty of the current risk transaction is a merchant and the merchant has more than a second preset number of risk transactions, the risk model is determined to be a marketing model; if the transaction device of the current risk transaction is a risk device, the risk model is determined to be a black device model, so as to determine at least one risk model corresponding to the current risk transaction.

[0064] In the embodiment of the present invention, it is achieved to determine the risk model of the current risk transaction according to the risk data of the current risk transaction.

[0065] Step 130: Input the risk channel and risk model of the current risk transaction into a pre-trained risk processing model, so that the risk processing model determines the processing method of the current risk transaction.

[0066] Among them, the risk handling model is obtained by training the decision tree model based on the risk model, risk channel and handling method of historical risk transactions.

[0067] There is at least one processing method corresponding to the same risk channel. For example, the processing methods corresponding to card risk include card stop payment, card reduction and account locking. In addition, the risk model also affects the choice of processing method. Therefore, the processing method of the current risk transaction can be determined based on the risk channel and risk model of the current risk transaction.

[0068] Specifically, the risk channel and risk model of the current risk transaction may be input into a pre-trained risk processing model, so that the risk processing model determines the processing method of the current risk transaction.

[0069] In one embodiment, the risk management model training process includes:

[0070] The risk model, risk channel and processing method of historical risk transactions are obtained, and the risk model and risk channel of each of the historical risk transactions are used as risk features, and the processing method of each of the historical risk transactions is used as label information to construct a training set; the target feature is determined according to the evaluation index of each of the risk features in the sample set corresponding to the current node in the decision tree, and the sample set corresponding to the left node and the sample set corresponding to the right node of the current node are determined according to the target feature and the feature value corresponding to the target feature, wherein the sample set corresponding to the root node of the decision tree is the training set, and the sample sets corresponding to the internal nodes are all subsets of the training set; the left node and the right node are respectively used as the current node, and the execution is returned to determine the target feature according to the evaluation index of each of the risk features in the sample set corresponding to the current node in the decision tree until the second preset end condition is met, and the risk processing model is obtained.

[0071] Specifically, first, the historical risk transactions that have been successfully processed in the historical time period and the risk channels and processing methods of each historical risk transaction can be obtained from the risk monitoring system. The risk channels and processing methods of each historical risk transaction are determined by the risk processing personnel and input into the risk monitoring system. Secondly, the risk model and risk channel of each historical risk transaction can be used as risk features, and the processing methods of each historical risk transaction can be used as label information to construct a training set. Then, starting from the root node of the decision tree, the root node can be used as the current node, and the target feature can be determined according to the evaluation index of each risk feature in the sample set corresponding to the current node, and the sample set corresponding to the left node of the current node and the sample set corresponding to the right node can be determined according to the target feature and the feature value corresponding to the target feature. After determining the sample set corresponding to the left node of the current node and the sample set corresponding to the right node, the left node and the right node can be used as the current node respectively, and the target feature can be determined according to the evaluation index of each risk feature in the sample set corresponding to the current node until the second preset end condition is met, and the risk processing model is obtained.

[0072] The second preset convergence condition may be that the processing method of each historical risk transaction in the sample set corresponding to the left node is consistent, the processing method of each historical risk transaction in the sample set corresponding to the right node is consistent, there is no risk feature for splitting the sample set, and any one of the predefined tree depths is reached.

[0073] It should be noted that the sample set corresponding to the root node of the decision tree is the training set, and the sample sets corresponding to the internal nodes are all subsets of the training set.

[0074] In addition, the evaluation index of each risk feature can be determined according to the information gain, Gini index or chi-square test of each risk feature in the sample set corresponding to the current node, and the target feature can be determined according to the evaluation index of each risk feature.

[0075] Therefore, after the risk channels and risk models of the current risk transaction are input into the risk processing model, the risk processing model can use the risk channels and risk models of the current risk transaction as risk features, and divide the current risk transaction based on each node in turn until the current risk transaction is divided into at least one sub-node. Then, the processing methods corresponding to these sub-nodes can be determined as the processing methods corresponding to the current risk transaction, thereby realizing the determination of the processing method corresponding to the current risk transaction based on the risk processing model.

[0076] It should also be noted that both the risk classification model and the risk processing model can be distributed in distributed microservers, and the microservers distributed by the risk classification model can be connected to the microservers distributed by the risk processing model, that is, the risk channel of the current risk transaction output by the risk classification model can be directly input into the risk processing model, and the risk processing model can combine the risk channel of the current risk transaction output by the risk classification model and the risk model of the current risk transaction to output the processing method of the current risk transaction. The risk classification model and the risk processing model interact through distributed microservices, and the time taken for the risk transaction from the occurrence to the determination of the risk channel and the processing method is in seconds, which greatly improves the timeliness of risk circulation and prevents risk spillover.

[0077] Of course, after determining the processing method of the current risk transaction, the current risk transaction can be processed according to the processing method to achieve real-time automatic risk processing of the current risk transaction.

[0078] In an embodiment of the present invention, the risk channel and risk model of the current risk transaction are input into the risk processing model obtained by training the decision tree with the risk model, risk channel and processing method of the historical risk transaction, so that the risk processing model determines the processing method of the current risk transaction according to the risk channel and risk model of the current risk transaction, thereby realizing rapid and accurate determination of the processing method of the risk transaction, and achieving the optimization purpose of eliminating manual work and improving effectiveness.

[0079] The method for processing risk transactions provided by an embodiment of the present invention includes: inputting the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions; determining a risk model based on the risk data of the current risk transaction; inputting the risk channel and risk model of the current risk transaction into a pre-trained risk processing model, so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions. The above technical scheme, after acquiring the risk transaction data of the current risk transaction, can input the risk data of the current risk transaction into the risk classification model obtained by training the semi-supervised generative adversarial network with the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions, and determine the risk channel of the current risk transaction based on the risk classification model, so as to achieve rapid and accurate determination of the risk channel of the risk transaction, and can also determine the risk model of the current risk transaction based on the risk data of the current risk transaction, and then input the risk channel and risk model of the current risk transaction into the risk processing model obtained by training the decision tree with the risk model, risk channel and processing method of historical risk transactions, and determine the processing method of the current risk transaction based on the risk processing model, so as to achieve rapid and accurate determination of the processing method of the risk transaction, and can process the current risk transaction according to the processing method, so as to achieve real-time automatic risk processing of the current risk transaction.

[0080] Figure 2 This is a schematic diagram of the structure of a risk transaction processing device provided by an embodiment of the present invention, which can be applied to situations where risk transactions need to be processed accurately and quickly. The device can be implemented by software and / or hardware and is generally integrated in an electronic device, such as a computer device.

[0081] like Figure 2 As shown, the device comprises:

[0082] A classification module 210, configured to input the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions;

[0083] A determination module 220, configured to determine a risk model according to the risk data of the current risky transaction;

[0084] The processing module 230 is used to input the risk channel and risk model of the current risk transaction into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

[0085] The risk transaction processing device provided in this embodiment inputs the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions; determines the risk model based on the risk data of the current risk transaction; inputs the risk channel and risk model of the current risk transaction into a pre-trained risk processing model, so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions. The above technical scheme, after acquiring the risk transaction data of the current risk transaction, can input the risk data of the current risk transaction into the risk classification model obtained by training the semi-supervised generative adversarial network with the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions, determine the risk channel of the current risk transaction based on the risk classification model, and realize rapid and accurate determination of the risk channel of the risk transaction, and can also determine the risk model of the current risk transaction based on the risk data of the current risk transaction, and then input the risk channel and risk model of the current risk transaction into the risk processing model obtained by training the decision tree with the risk model, risk channel and processing method of historical risk transactions, and determine the processing method of the current risk transaction based on the risk processing model, and realize rapid and accurate determination of the processing method of the risk transaction.

[0086] In one implementation, the risk channel is card risk, customer risk, merchant risk or equipment risk.

[0087] Based on the above embodiment, the device further includes:

[0088] The first training module is used to obtain the risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions; the risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions are used as training data to train the semi-supervised generative adversarial network until the first preset end condition is met to obtain the risk classification model.

[0089] In one embodiment, the risk model is at least one of a long-term inactive account model, a money laundering model, a low credit model, a one-person-multiple-account model, a marketing model, and a black device model.

[0090] Based on the above embodiment, the determination module 220 is specifically configured to:

[0091] Determining whether the current risk transaction satisfies the risk conditions of each risk model according to the risk data of the current risk transaction;

[0092] In the case where it is determined that the current risk transaction satisfies the risk condition of the risk model, the risk model is determined as the risk model of the current risk transaction.

[0093] Based on the above embodiment, the device further includes:

[0094] The second training module is used to obtain the risk model, risk channel and processing method of historical risk transactions, take the risk model and risk channel of each historical risk transaction as risk features, and take the processing method of each historical risk transaction as label information to construct a training set; determine the target feature according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, and determine the sample set corresponding to the left node and the sample set corresponding to the right node of the current node according to the target feature and the feature value corresponding to the target feature, wherein the sample set corresponding to the root node of the decision tree is the training set, and the sample sets corresponding to the internal nodes are all subsets of the training set; take the left node and the right node as the current node respectively, return to execute the target feature according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, until the second preset end condition is met, and the risk processing model is obtained.

[0095] The risk transaction processing device provided in the embodiment of the present invention can execute the risk transaction processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the risk transaction processing method.

[0096] It is worth noting that in the embodiment of the above-mentioned risk transaction processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0097] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3 A block diagram of an exemplary electronic device 3 suitable for implementing embodiments of the present invention is shown. Figure 3The electronic device 3 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0098] like Figure 3 As shown, the electronic device 3 is in the form of a general-purpose computing electronic device. The components of the electronic device 3 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0099] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0100] The electronic device 3 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 3, including volatile and non-volatile media, removable and non-removable media.

[0101] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 3 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 3 not shown, usually called a "hard drive"). Although Figure 3 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0102] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0103] The electronic device 3 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the electronic device 3, and / or any device that enables the electronic device 3 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the electronic device 3 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. Figure 3 As shown, the network adapter 20 communicates with other modules of the electronic device 3 via the bus 18. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 3, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0104] The processing unit 16 executes various functional applications and page displays by running the programs stored in the system memory 28, for example, implementing the risk transaction processing method provided by the embodiment of the present invention, the method comprising:

[0105] Inputting the risk data of the current risk transaction into a pre-trained risk classification model so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions;

[0106] Determining a risk model based on the risk data of the current risk transaction;

[0107] The risk channel and risk model of the current risk transaction are input into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

[0108] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the risk transaction processing method provided by any embodiment of the present invention.

[0109] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, a method for processing risky transactions provided by an embodiment of the present invention is implemented. The method includes:

[0110] Inputting the risk data of the current risk transaction into a pre-trained risk classification model so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions;

[0111] Determining a risk model based on the risk data of the current risk transaction;

[0112] The risk channel and risk model of the current risk transaction are input into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

[0113] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0114] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0115] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0116] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0117] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0118] In addition, the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0119] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for processing risky transactions, characterized in that: include: Inputting the risk data of the current risk transaction into a pre-trained risk classification model so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions; Determining a risk model based on the risk data of the current risk transaction; The risk channel and risk model of the current risk transaction are input into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

2. The method for processing risky transactions according to claim 1, characterized in that: The risk channel is card risk, customer risk, merchant risk or equipment risk.

3. The method for processing risky transactions according to claim 1, characterized in that: The training process of the risk classification model includes: Obtain risk data and risk channels of historical risk transactions and risk data of the real-time risk transactions; The risk data and risk channels of the historical risk transactions and the risk data of the real-time risk transactions are used as training data to train the semi-supervised generative adversarial network until a first preset end condition is met, thereby obtaining the risk classification model.

4. The method for processing risky transactions according to claim 1, characterized in that: The risk model is at least one of a long-term inactive account model, a money laundering model, a low credit model, a one-person-multiple-account model, a marketing model, and a black device model.

5. The method for processing risky transactions according to claim 4, characterized in that: Determining a risk model according to the risk data of the current risk transaction includes: Determining whether the current risk transaction satisfies the risk conditions of each risk model according to the risk data of the current risk transaction; In the case where it is determined that the current risk transaction satisfies the risk condition of the risk model, the risk model is determined as the risk model of the current risk transaction.

6. The method for processing risky transactions according to claim 1, characterized in that: The training process of the risk treatment model includes: Obtain risk models, risk channels, and processing methods of historical risk transactions, use the risk models and risk channels of each of the historical risk transactions as risk features, and use the processing methods of each of the historical risk transactions as label information to construct a training set; Determine the target feature according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, and determine the sample set corresponding to the left node and the sample set corresponding to the right node of the current node according to the target feature and the feature value corresponding to the target feature, wherein the sample set corresponding to the root node of the decision tree is the training set, and the sample sets corresponding to the internal nodes are all subsets of the training set; The left node and the right node are respectively used as the current node, and the target feature is determined according to the evaluation index of each risk feature in the sample set corresponding to the current node in the decision tree, until the second preset end condition is met, so as to obtain the risk processing model.

7. A risk transaction processing device, characterized in that: include: A classification module, used for inputting the risk data of the current risk transaction into a pre-trained risk classification model, so that the risk classification model determines the risk channel of the current risk transaction, wherein the risk classification model is obtained by training a semi-supervised generative adversarial network based on the risk data and risk channels of historical risk transactions and the risk data of real-time risk transactions; A determination module, configured to determine a risk model according to the risk data of the current risky transaction; A processing module is used to input the risk channel and risk model of the current risk transaction into a pre-trained risk processing model so that the risk processing model determines the processing method of the current risk transaction, wherein the risk processing model is obtained by training a decision tree model based on the risk model, risk channel and processing method of historical risk transactions.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for processing risk transactions as described in any one of claims 1-6.

9. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to execute the risk transaction processing method as described in any one of claims 1-6 when executed by a computer processor.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the method for processing risk transactions according to any one of claims 1 to 6.

Images