WO2019144808A1

WO2019144808A1 - Method and apparatus for determining false resource transfer, method and apparatus for determining false trading, and electronic device

Info

Publication number: WO2019144808A1
Application number: PCT/CN2019/071113
Authority: WO
Inventors: 程羽; 陈弢
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2018-01-23
Filing date: 2019-01-10
Publication date: 2019-08-01
Also published as: TW201933226A; CN108492112B; CN108492112A

Abstract

A method and apparatus for determining a false resource transfer, a method and apparatus for determining false trading, and an electronic device, for use in solving the problem of optimization of method for determining false trading. The method comprises: obtaining past resource transfer data of a resource transferee within a first predetermined time period before a resource transfer to be verified, and behavior data of the resource transferee within a second predetermined time period before the resource transfer to be verified (110); and determining whether the resource transfer to be verified is a false resource transfer according to the past resource transfer data, the behavior data, and a false resource transfer model (120), wherein the false resource transfer model is trained on the basis of past resource transfer training data and corresponding behavior training data.

Description

Method, device and electronic device for determining false resource transfer and false transaction

Technical field

The present application relates to the field of computer technologies, and in particular, to a method, device, and electronic device for determining a false resource transfer and a fraudulent transaction.

Background technique

At present, with the rapid development of Internet technology, e-commerce platforms such as Jingdong and Taobao have gradually become an indispensable part of people's daily lives. When a buyer wants to purchase an item, he can browse through the e-commerce platform to browse the item he wants to purchase, and select an item to complete the transaction. Most buyers choose a certain product to compare a number of similar products, usually based on information from other buyers on the product, seller's business, and seller's credit points to determine whether a product is worth buying.

However, in order to obtain a good evaluation of their merchants or a certain commodity, sellers on the e-commerce platform often obtain improper sales such as merchandise sales, merchant ratings, credit scores, etc. through improper methods (such as flash transactions and other false transactions). In order to make the buyer make a wrong judgment when purchasing the goods, thereby hindering the buyer's rights.

Therefore, how to accurately and effectively identify false transactions has become one of the most important issues to be solved in e-commerce platforms.

Summary of the invention

The embodiment of the present application provides a method, a device, and an electronic device for determining a false resource transfer and a false transaction, so as to solve the problem that the method for determining a false transaction in the prior art is not optimized.

To solve the above technical problem, the embodiment of the present application is implemented as follows:

In the first aspect, a method for determining the transfer of false resources is proposed, including:

Obtaining, by the resource transferee, historical resource transfer data within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transferee within a second predetermined time period before the resource to be verified is transferred;

Determining, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

The fake resource transfer model is trained based on historical resource transfer training data and corresponding behavior training data.

In the second aspect, a method for determining a false transaction is proposed, including:

Obtaining historical transaction data of the transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical resource transfer data, the behavior data, and a fraudulent transaction model;

The false transaction model is trained based on historical transaction training data and corresponding behavioral training data.

In a third aspect, an apparatus for determining a false resource transfer is provided, including:

An acquiring unit, acquiring historical resource transfer data of the resource transfer party within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transfer party within a second predetermined time period before the resource to be verified is transferred ;

The determining unit determines, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

In a fourth aspect, an apparatus for determining a fraudulent transaction is provided, comprising:

Obtaining a unit, acquiring historical transaction data of a transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

a determining unit, determining, according to the historical resource transfer data, the behavior data, and a fraudulent transaction model, whether the transaction to be verified is a fraudulent transaction;

In a fifth aspect, an electronic device is provided, the electronic device comprising:

Processor;

A memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:

In a sixth aspect, a computer readable storage medium is provided, the computer readable storage medium storing one or more programs, when the one or more programs are executed by an electronic device including a plurality of applications, The electronic device performs the following operations:

In a seventh aspect, an electronic device is provided, the electronic device comprising:

Processor;

In an eighth aspect, a computer readable storage medium is presented, the computer readable storage medium storing one or more programs that, when executed by an electronic device comprising a plurality of applications, cause The electronic device performs the following operations:

The foregoing technical solutions adopt the above technical solutions to achieve at least the following technical effects:

Obtaining the historical resource transfer data in the first predetermined time period before the resource transfer party is transferred, and the behavior data in the second predetermined time period before the resource transfer party transfers the resource to be verified, based on the acquired historical resource Transfer data, behavioral data and false resource transfer model, determine whether the resource transfer to be verified is a false resource transfer, not only consider the historical resource transfer data of the resource transferee, but also use the behavior data before the resource transfer as the discriminant to identify the false resource. The basis of the transfer improves the accuracy of discriminating the transfer of false resources and achieves the purpose of optimizing the identification of the transfer of false resources.

Obtaining historical transaction data within a first predetermined time period before the transaction to be verified by the transaction payer and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified, based on the acquired historical transaction data, behavior data, and The false transaction model determines whether the transaction to be verified is a false transaction. It not only considers the historical transaction data of the transaction payer, that is, the buyer, but also uses the behavior data before the transaction as the basis for discriminating the false transaction, and improves the discrimination of the false transaction. The accuracy of the goal is to optimize the identification of false transactions.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the present application, and are intended to be a part of this application. In the drawing:

FIG. 1 is a schematic flowchart of an implementation method for determining a false resource transfer according to an embodiment of the present disclosure;

2 is a schematic diagram of an implementation process of a method for determining a fake transaction according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a data processing process of a method for determining a fake transaction according to an embodiment of the present disclosure; FIG.

4 is a schematic diagram of a process of performing vectorization preprocessing of behavior data in a method for determining a false transaction according to an embodiment of the present disclosure;

FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;

FIG. 6 is a schematic structural diagram of another electronic device according to an embodiment of the present disclosure;

FIG. 7 is a schematic structural diagram of an apparatus for determining a false resource transfer according to an embodiment of the present disclosure;

FIG. 8 is a schematic structural diagram of an apparatus for determining a fake transaction according to an embodiment of the present disclosure.

Detailed ways

The technical solutions of the present application will be clearly and completely described in the following with reference to the specific embodiments of the present application and the corresponding drawings. It is apparent that the described embodiments are only a part of the embodiments of the present application, and not all of them. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

In order to solve the problem that the method for determining a false transaction in the prior art is not optimized, the embodiment of the present specification provides a method for determining a false resource transfer. The execution subject of the method for determining the false resource transfer provided by the embodiment of the present disclosure may be, but is not limited to, a server, a personal computer, or the like, which can be configured to perform at least one of the method terminals provided by the embodiments of the present invention.

For convenience of description, an embodiment of the method will be described below by taking an example of the execution of the method as a server capable of executing the method. It can be understood that the execution subject of the method is only an exemplary description of the server, and should not be construed as limiting the method.

Specifically, a schematic flowchart for implementing a method for determining a false resource transfer provided by one or more embodiments of the present disclosure is as shown in FIG. 1 , and includes:

Step 110: Obtain historical resource transfer data of the resource transfer party within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transfer party within a second predetermined time period before the resource to be verified is transferred;

The historical resource transfer data includes at least one of the following: the number of historical resource transfers, the historical resource transfer quota, and the number of resource recipients involved in the historical resource transfer. In practical applications, the resource transfer may be, for example, a transaction, and the historical resource transfer data may specifically include: a historical transaction number, a historical transaction amount, a transaction transaction involved in the historical transaction, that is, a seller or a merchant, and the single transaction is the largest. Value, single transaction minimum, single transaction average, daily trading maximum, daily trading minimum, daily trading average, etc., can be used as the transaction payer, that is, the historical transaction data of the buyer before the transaction. .

The behavior data includes at least one of the following: the information of the resource receiver browsed by the resource transferee, the browsing duration, and the browsed resource information, and the information of the resource receiver includes at least the credit value, the resource category, and the resource value of the resource receiver. Distribution, health. In practical applications, the behavior data may specifically include: information of the seller or the merchant that the buyer browsed before completing the transaction, and the length of time that each seller or merchant stays while browsing, and the products browsed in each merchant. Information (such as the name of the product, price, etc.), where the seller or merchant's information includes the credit value of the seller or merchant, the category of the commodity in the merchant, the price distribution of the commodity, and the recent complaints of the merchant (such as within a month) The number of times, and so on.

In addition, the first predetermined time period may be a certain period of time before the resource transfer party, such as a period of three months, one month or one week before the resource transfer, the second reservation The time period may be a few days or a day or a few hours before the resource transfer party, and the first predetermined time period and the second predetermined time period may be set according to actual needs. The embodiment or embodiments do not specifically limit this.

It should be understood that in practical applications, in the non-fake resource transfer, most buyers are prepared to browse a certain type of goods on the e-commerce platform, often browsing the sellers or merchants related to the goods, and browsing these products and determining When you want to buy, you will often check the price of the item, the introduction of the item, the evaluation of the item by other buyers, and the credit value of the seller or merchant who sells the item, and so on. In the case of false resource transfer, the resource transfer party often raises the credit value of the seller or the merchant for the improper reception of the resource recipients, that is, some sellers or merchants, to complete the transfer of a certain false resource. At the same time, it is often not possible to browse the sellers or merchants associated with the goods in advance, or deliberately browse some sellers or merchants related to the goods for the purpose of hiding the false transactions.

Although there are similar behavioral data in the transfer of false resources, the motivation and non-fake resource transfer are not the same in essence. This can be reflected in a series of behavior data before the resource transfer party, such as false resources. There is a big difference in the browsing time of the resource transferee in the transfer and non-fake resource transfer when browsing the products in each seller or merchant, and the information content of the viewed seller or the product. Based on this point, one or more embodiments of the present specification not only consider the historical resource transfer data of the resource transfer party within the first predetermined time period before the resource transfer to be verified, but also the resource transfer party in the resource transfer to be verified. The behavior data in the first two predetermined time periods, thereby improving the accuracy of identifying the transfer of false resources, and further safeguarding the rights of other resource transfer parties, that is, buyers.

Step 120: Determine, according to the historical resource transfer data, the behavior data, and the false resource transfer model, whether the resource transfer to be verified is a false resource transfer; wherein the false resource transfer model is trained based on the historical resource transfer training data and the corresponding behavior training data.

Specifically, based on the historical resource transfer data, the behavior data, and the false resource transfer model, determining whether the resource transfer to be verified is a false resource transfer, the historical resource transfer characteristic data may be first determined according to the historical resource transfer data, and according to the behavior data, The behavior characteristic data is determined. Finally, the historical resource transfer characteristic data, the behavior characteristic data, and the false resource transfer model can be used to determine whether the resource transfer to be verified is a false resource transfer.

As shown in FIG. 2 , the historical resource transfer data, the behavior data, and the false resource transfer model provided by one or more embodiments of the present specification determine whether the resource transfer to be verified is a schematic diagram of a false resource transfer process. The historical resource transfer data includes the buyer information and the like data shown in FIG. 2, and the buyer information may include information such as the number of resource transfers and the resource transfer quota of the buyer within the first predetermined time period before the resource transfer. Since the historical resource transfer data is the resource transfer party, that is, the historical resource transfer data of the buyer before the resource transfer, it will not change with the change of the behavior of the resource transfer party, so the history may also be used. The resource transfer data is called static data; the behavior data includes the business information shown in Figure 2, the trader's transaction history, and the browsing log. Since the behavior data changes with the buyer's behavior, the behavior data can also be used. Called dynamic data.

After the static data and the dynamic data are acquired, the historical resource transfer feature data may be determined according to the static data, and the behavior feature data is determined according to the dynamic data. Since the dynamic data may include data that cannot be directly represented by a vector, such as a business address, when determining the behavior characteristic data according to the dynamic data, it is also necessary to encode the data in the dynamic data that cannot be directly represented by the vector. This is the vectorization preprocessing described below. After the feature data of the static data and the dynamic data are respectively determined, the two feature data can be spliced based on the two feature data and the fake resource transfer model, and the second classifier algorithm is used to determine whether the resource transfer is performed. Transfer for false resources.

It should be understood that, since the above behavior data may include data that the resource receiver, that is, the ID of the seller or the merchant, cannot directly represent by the vector, in order to facilitate the processing of the behavior data, according to the behavior data, the behavior characteristic data is determined. Vectorization preprocessing of data in the behavior data that cannot be directly characterized by vectors can be performed first to convert non-vector data in the behavior data into vector data; and since the behavior data includes data of multiple feature dimensions, in order to unify these features Dimensional data is dimensioned, so the vectorized preprocessed behavior data is also normalized to obtain behavioral feature data. Among them, the method of vectorizing the behavior data may adopt a tool for converting a string into a vector form such as a word2vector algorithm, an embedding algorithm, and the like.

As shown in FIG. 3, a schematic diagram of a process for processing behavior data provided by one or more embodiments of the present specification, in FIG. 3, data that cannot be directly represented by a vector in behavior data can be converted into a vectorized preprocessing to The vector form, as shown in the "Click to Browse ID" data in Figure 3, is the ID of a merchant that the resource transferee viewed before the resource transfer. Since the data is "00N5789Y218", that is, a string form, it cannot be directly represented by a vector. In order to facilitate the processing of the behavior data, the string can be converted into a vector form by the word2vector algorithm, and the obtained "browsing merchant information" and "browsing details" are represented by vectors, and these are The behavior data characterized as a vector form is spliced into a multi-dimensional vector, and the unified dimension is processed by normalization.

Before deciding whether the resource transfer to be verified is a false resource transfer based on historical resource transfer data, behavior data and false resource transfer model, it can be trained through two modes: supervised two-class training mode and unsupervised two-class training mode. False resource transfer model:

(1) Supervised two-class training method

Firstly, the historical resource transfer training data is normalized to obtain the corresponding historical resource transfer feature data; then the data in the behavior training data that cannot be directly represented by the vector is vectorized and preprocessed; then, after vectorization preprocessing The behavioral training data is normalized to obtain the corresponding behavioral feature data. Finally, the historical resource transfer characteristic data, the behavioral feature data and the corresponding resource transfer type are input as inputs, and the false resource transfer model is trained, wherein the resource transfer type Includes non-fake resource transfers and non-fake resource transfers.

In this manner, since the resource transfer type includes non-false resource transfer and non-false resource transfer, the historical resource transfer training data and the corresponding behavior training data include: a plurality of non-false resource transfer resource transfer party historical resources The transfer data and the corresponding behavior data, and the historical resource transfer data and the corresponding behavior data of the resource transfer party of the multiple false resource transfer; the false resource transfer model is trained based on the historical resource transfer training data and the corresponding behavior training data. The process can include:

In step i, the historical resource transfer data of the resource transfer party of the non-fake resource transfer may be normalized to obtain the corresponding historical resource transfer feature data of the non-fake resource transfer, and the resources transferred to the multiple false resources The historical resource transfer data of the transferee is normalized to obtain the corresponding historical resource transfer feature data of the multiple false resource transfer, wherein the historical resource transfer data of the resource transfer party of the multiple non-false resource transfer is the resource transfer The historical resource transfer data of the first predetermined time period before the corresponding non-fake resource transfer, and the historical resource transfer data of the resource transfer party of the multiple false resource transfer are the resource transfer party before the corresponding false resource transfer Historical resource transfer data within a first predetermined time period;

Step ii: performing vectorization preprocessing on data that cannot be directly represented by the vector in the corresponding behavior data in the multiple non-fake resource transitions; performing vector on the data that cannot be directly represented by the vector in the corresponding behavior data in the multiple false resource transitions The pre-processing, wherein the corresponding behavior data in the multiple non-fake resource transfers is the behavior data of the resource transfer party in the second predetermined time period before the corresponding non-fake resource transfer, and the corresponding behavior in the multiple false resource transfers The data is behavior data of the resource transfer party within a second predetermined time period before the corresponding false resource transfer;

Step iii: normalize the corresponding behavior data in the plurality of non-fake resource transitions after the vectorization pre-processing to obtain corresponding behavior characteristic data in the plurality of non-false resource transitions; The corresponding behavior data in the multiple false resource transitions after processing is normalized to obtain corresponding behavior characteristic data in multiple false resource transitions;

Step iv: training a false resource transfer model based on historical resource transfer feature data and corresponding behavior feature data of multiple non-fake resource transfers, and historical resource transfer feature data and corresponding behavior feature data of multiple false resource transfers. In practical applications, the fake resource transfer module may be based on historical resource transfer feature data and corresponding behavior feature data of multiple non-false resource transfers, and historical resource transfer feature data and corresponding behavior feature data of multiple false resource transfers. The classifier is trained, and the specific training method can refer to the related model training method in the prior art, and will not be described again.

(2) Unsupervised two-class training method

Firstly, the historical resource transfer training data is normalized to obtain the corresponding historical resource transfer feature data. Then, the data that can not be directly represented by the vector in the behavior training data is vectorized and preprocessed, and the vectorized preprocessed The behavioral training data is normalized and the corresponding behavioral feature data is obtained. Finally, the historical resource transfer characteristic data and behavioral feature data are taken as input, and the false resource transfer model is obtained by clustering training according to the two classifications. The specific training method can be referred to. The related model training methods in the prior art are not described again.

4 is a schematic flowchart of an implementation of a method for determining a fraudulent transaction provided by an embodiment of the present specification, including:

Step 210: Obtain historical transaction data of the transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

The historical transaction data includes at least one of the following: a historical transaction number, a historical transaction amount, and a number of transaction recipients involved in the historical transaction; and the behavior data includes at least one of the following: the transaction recipient's information of the transaction recipient browsing, The browsing time, the viewed product information, and the transaction recipient's information include at least the transaction recipient's credit value, product category, commodity price distribution, and health level.

Step 220: Determine whether the transaction to be verified is a false transaction based on historical resource transfer data, behavior data, and a false transaction model; the fake transaction model is trained based on historical transaction training data and corresponding behavior training data.

Specifically, based on the historical transaction data, the behavior data, and the false transaction model, determining whether the transaction to be verified is a fraudulent transaction, the historical transaction characteristic data may be first determined according to the historical transaction data; and then, the behavior characteristic data is determined according to the behavior data; Finally, based on historical transaction characteristic data, behavioral feature data and false transaction model, it is determined whether the transaction to be verified is a false transaction.

Optionally, according to the behavior data, the behavior characteristic data is determined. Specifically, the data that cannot be directly represented by the vector in the behavior data may be subjected to vectorization preprocessing; then, the behavioral data after the vectorization preprocessing is normalized. Process to get behavioral feature data.

Before determining whether the transaction to be verified is a false transaction based on historical transaction data, behavior data and false transaction model, the false transaction model can be trained through two modes: supervised two-class training mode and unsupervised two-class training mode:

(1) Supervised two-class training method

Firstly, the historical transaction training data is normalized to obtain the corresponding historical transaction characteristic data; then the data in the behavior training data that cannot be directly represented by the vector is vectorized and preprocessed; then, the vectorized preprocessed behavior is performed. The training data is normalized to obtain corresponding behavioral feature data. Finally, historical transaction characteristic data, behavioral feature data and corresponding transaction types are input as inputs, and a false transaction model is trained, wherein the transaction types include non-false transactions and non-transactions. a.

(2) Unsupervised two-class training method

Firstly, the historical transaction training data is normalized to obtain the corresponding historical transaction feature data. Then, the data in the behavior training data that cannot be directly represented by the vector is vectorized and preprocessed, and the vectorized preconditioned behavior training is performed. The data is normalized to obtain corresponding behavioral feature data; finally, historical transaction characteristic data and behavioral feature data are taken as inputs, and clustering training is performed according to the two classifications to obtain the false transaction model.

For a specific implementation of the steps related to the embodiment shown in FIG. 4, reference may be made to the specific implementation of the corresponding steps in the embodiment shown in FIG. 1 to FIG. 3. One or more embodiments of the present specification are not described herein again.

FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present specification. Referring to FIG. 5, at the hardware level, the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. Of course, the electronic device may also include hardware required for other services.

The processor, the network interface, and the memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended) Industry Standard Architecture, extending the industry standard structure) bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 5, but it does not mean that there is only one bus or one type of bus.

Memory for storing programs. In particular, the program can include program code, the program code including computer operating instructions. The memory can include both memory and non-volatile memory and provides instructions and data to the processor.

The processor reads the corresponding computer program from the non-volatile memory into memory and then runs, forming a device at the logical level to determine the transfer of the fake resource. The processor executes the program stored in the memory and is specifically used to perform the following operations:

The method for determining a false resource transfer disclosed in the embodiment shown in FIG. 1 of the present specification may be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processor (DSP), dedicated integration. Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component. The methods, steps, and logic blocks disclosed in one or more embodiments of the specification can be implemented or executed. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like. The steps of a method disclosed in connection with one or more embodiments of the present specification can be directly embodied as a hardware decoding processor or a combination of hardware and software modules in a decoding processor. The software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.

The electronic device can also perform the method for determining the false resource transfer in FIG. 1 , and the description is not repeated herein.

Of course, in addition to the software implementation, the electronic device of the present specification does not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution body of the following processing flow is not limited to each logical unit. It can also be hardware or logic.

FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present specification. Referring to FIG. 6, at the hardware level, the electronic device includes a processor, optionally including an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. Of course, the electronic device may also include hardware required for other services.

The processor, the network interface, and the memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended) Industry Standard Architecture, extending the industry standard structure) bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 6, but it does not mean that there is only one bus or one type of bus.

The processor reads the corresponding computer program from the non-volatile memory into memory and then runs, forming a device at the logical level to determine the fraudulent transaction. The processor executes the program stored in the memory and is specifically used to perform the following operations:

The method for determining a fraudulent transaction as disclosed in the embodiment shown in FIG. 4 of the present specification may be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processor (DSP), dedicated integration. Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component. The methods, steps, and logic blocks disclosed in one or more embodiments of the specification can be implemented or executed. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like. The steps of a method disclosed in connection with one or more embodiments of the present specification can be directly embodied as a hardware decoding processor or a combination of hardware and software modules in a decoding processor. The software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.

The electronic device can also perform the method for determining a fraudulent transaction of FIG. 4, and the description is not repeated herein.

FIG. 7 is a schematic structural diagram of an apparatus 700 for determining a false resource transfer provided by the present specification. Referring to FIG. 7, in a software implementation, the apparatus 700 for determining a false resource transfer may include an obtaining unit 701 and a determining unit 702, where:

The obtaining unit 701 is configured to acquire historical resource transfer data of the resource transfer party within a first predetermined time period before the resource to be verified is transferred, and behavior of the resource transfer party within a second predetermined time period before the resource to be verified is transferred. data;

The determining unit 702 determines, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

In an embodiment, the determining unit 702,

Determining historical resource transfer characteristic data according to the historical resource transfer data;

Determining behavior characteristic data according to the behavior data;

And determining, according to the historical resource transfer feature data, the behavior feature data, and the fake resource transfer model, whether the resource transfer to be verified is a fake resource transfer.

In an embodiment, the determining unit 702,

Performing vectorization preprocessing on data in the behavior data that cannot be directly characterized by vectors;

The behavior data subjected to the vectorization preprocessing is normalized to obtain the behavior characteristic data.

In an implementation manner, before the determining unit 702 determines, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer, the apparatus further includes :

The first processing unit 703 performs normalization processing on the historical resource transfer training data to obtain corresponding historical resource transfer feature data;

The second processing unit 704 performs vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data;

The third processing unit 705 performs normalization processing on the behavior training data after the vectorization preprocessing to obtain corresponding behavior feature data;

The first training unit 706, by using the historical resource transfer feature data, the behavior feature data, and the corresponding resource transfer type as an input, training the obtained false resource transfer model, wherein the resource transfer type includes a fake resource transfer and Non-false resource transfer.

The fourth processing unit 707 performs normalization processing on the historical resource transfer training data to obtain corresponding historical resource transfer feature data;

The fifth processing unit 708 performs vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data, and normalizes the behavior training data after the vectorization preprocessing to obtain a corresponding behavior. Characteristic data

The second training unit 709 takes the historical resource transfer feature data and the behavior feature data as input, and performs cluster training according to the two classifications to obtain the false resource transfer model.

In an embodiment, the historical resource transfer data includes at least one of the following:

The number of historical resource transfers, historical resource transfer quotas, and the number of resource recipients involved in historical resource transfers.

In an embodiment, the behavior data includes at least one of the following:

The information of the resource receiver of the resource transfer party, the browsing duration, and the browsed resource information, and the information of the resource receiver includes at least the credit value, resource category, resource value distribution, and health level of the resource receiver.

The device 700 for determining the virtual resource transfer can implement the method of the method embodiment of FIG. 1 to FIG. 3 . For details, refer to the method for determining the false resource transfer in the embodiment shown in FIG. 1 , and details are not described herein again.

FIG. 8 is a schematic structural diagram of an apparatus 800 for determining a fraudulent transaction provided by the present specification. Referring to FIG. 8, in a software implementation, the apparatus 800 for determining a fraudulent transaction may include an obtaining unit 801 and a determining unit 802, where:

In an embodiment, the determining unit 802,

Determining whether the transaction to be verified is a fraudulent transaction based on the historical transaction data, the behavior data, and a fraudulent transaction model, including:

Determining historical transaction characteristic data according to the historical transaction data;

Determining behavior characteristic data according to the behavior data;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical transaction feature data, the behavior feature data, and the fraudulent transaction model.

In an embodiment, the determining unit 802,

In an embodiment, before the determining unit 802 determines, according to the historical transaction data, the behavior data, and the fraudulent transaction model, whether the transaction to be verified is a fraudulent transaction, the device further includes:

The first processing unit 803 performs normalization processing on the historical transaction training data to obtain corresponding historical transaction feature data;

The second processing unit 804 performs vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data;

The third processing unit 805 normalizes the behavior training data after the vectorization preprocessing to obtain corresponding behavior feature data;

The first training unit 806 receives the historical transaction feature data, the behavior feature data, and the corresponding transaction type as inputs, and trains the obtained false transaction model, wherein the transaction type includes a false transaction and a non-fake transaction.

The fourth processing unit 807 performs normalization processing on the historical transaction training data to obtain corresponding historical transaction feature data;

The fifth processing unit 808 performs vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data, and normalizes the behavior training data after the vectorization preprocessing to obtain a corresponding behavior. Characteristic data

The second training unit 809 takes the historical transaction feature data and the behavior feature data as input, and performs cluster training according to the two classifications to obtain the false transaction model.

In one embodiment, the historical transaction data includes at least one of the following:

The number of historical transactions, the historical transaction amount, and the number of transaction recipients involved in historical transactions.

In an embodiment, the behavior data includes at least one of the following:

The transaction recipient's information, browsing duration, and browsed resource information browsed by the transaction payer, and the transaction recipient's information includes at least the transaction recipient's credit value, product category, commodity value distribution, and health level.

The device 800 for determining a fraudulent transaction can implement the method of the method embodiment of FIG. 4 . For details, refer to the method for determining a false transaction in the embodiment shown in FIG. 4 , and details are not described herein again.

In conclusion, the above description is only the preferred embodiment of the present specification, and is not intended to limit the scope of the present specification. Any modifications, equivalent substitutions, improvements, etc. within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

The system, device, module or unit illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or A combination of any of these devices.

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

It is also to be understood that the terms "comprises" or "comprising" or "comprising" or any other variations are intended to encompass a non-exclusive inclusion, such that a process, method, article, Other elements not explicitly listed, or elements that are inherent to such a process, method, commodity, or equipment. An element defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device including the element.

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

Claims

A method for determining the transfer of a false resource, comprising:

Obtaining, by the resource transferee, historical resource transfer data within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transferee within a second predetermined time period before the resource to be verified is transferred;

Determining, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

The fake resource transfer model is trained based on historical resource transfer training data and corresponding behavior training data.
The method of claim 1

And determining, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer, including:

Determining historical resource transfer characteristic data according to the historical resource transfer data;

Determining behavior characteristic data according to the behavior data;

And determining, according to the historical resource transfer feature data, the behavior feature data, and the fake resource transfer model, whether the resource transfer to be verified is a fake resource transfer.
The method of claim 2, determining behavior characteristic data based on the behavior data, comprising:

Performing vectorization preprocessing on data in the behavior data that cannot be directly characterized by vectors;

The behavior data subjected to the vectorization preprocessing is normalized to obtain the behavior characteristic data.
The method of claim 1, before the determining whether the resource to be verified is a fake resource transfer based on the historical resource transfer data, the behavior data, and the fake resource transfer model, the method further includes:

Normalizing the historical resource transfer training data to obtain corresponding historical resource transfer feature data;

Performing vectorization preprocessing on data that cannot be directly characterized by vectors in the behavior training data;

Normalizing the behavior training data after the vectorization preprocessing to obtain corresponding behavior characteristic data;

The historical resource transfer feature data, the behavior feature data, and the corresponding resource transfer type are input as inputs, and the fake resource transfer model is trained, wherein the resource transfer type includes a fake resource transfer and a non-fake resource transfer.
The method of claim 1, before the determining whether the resource to be verified is a fake resource transfer based on the historical resource transfer data, the behavior data, and the fake resource transfer model, the method further includes:

Normalizing the historical resource transfer training data to obtain corresponding historical resource transfer feature data;

Performing vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data, and normalizing the behavior training data after the vectorization preprocessing to obtain corresponding behavior characteristic data;

Taking the historical resource transfer feature data and the behavior feature data as input, performing cluster training according to the two classifications to obtain the false resource transfer model.
A method according to any one of claims 1 to 5,

The historical resource transfer data includes at least one of the following:

The number of historical resource transfers, historical resource transfer quotas, and the number of resource recipients involved in historical resource transfers.
A method according to any one of claims 1 to 5,

The behavior data includes at least one of the following:

The information of the resource receiver of the resource transfer party, the browsing duration, and the browsed resource information, and the information of the resource receiver includes at least the credit value, resource category, resource value distribution, and health level of the resource receiver.
A method of determining a fraudulent transaction, including:

Obtaining historical transaction data of the transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical resource transfer data, the behavior data, and a fraudulent transaction model;

The false transaction model is trained based on historical transaction training data and corresponding behavioral training data.
The method of claim 8, determining whether the transaction to be verified is a fraudulent transaction based on the historical transaction data, the behavior data, and a fraudulent transaction model, comprising:

Determining historical transaction characteristic data according to the historical transaction data;

Determining behavior characteristic data according to the behavior data;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical transaction feature data, the behavior feature data, and the fraudulent transaction model.
The method of claim 9, determining behavior characteristic data based on the behavior data, comprising:

Performing vectorization preprocessing on data in the behavior data that cannot be directly characterized by vectors;

The behavior data subjected to the vectorization preprocessing is normalized to obtain the behavior characteristic data.
The method of claim 8, before determining whether the transaction to be verified is a fraudulent transaction, based on the historical transaction data, the behavior data, and the fraudulent transaction model, the method further comprising:

Normalizing the historical transaction training data to obtain corresponding historical transaction feature data;

Performing vectorization preprocessing on data that cannot be directly characterized by vectors in the behavior training data;

Normalizing the behavior training data after the vectorization preprocessing to obtain corresponding behavior characteristic data;

The historical transaction characteristic data, the behavior characteristic data, and the corresponding transaction type are input as inputs, and the false transaction model is trained, wherein the transaction type includes a false transaction and a non-false transaction.
The method of claim 8, before determining whether the transaction to be verified is a fraudulent transaction, based on the historical transaction data, the behavior data, and the fraudulent transaction model, the method further comprising:

Normalizing the historical transaction training data to obtain corresponding historical transaction feature data;

Performing vectorization preprocessing on the data that cannot be directly represented by the vector in the behavior training data, and normalizing the behavior training data after the vectorization preprocessing to obtain corresponding behavior characteristic data;

Taking the historical transaction feature data and the behavior feature data as input, performing cluster training according to the two classifications to obtain the false transaction model.
A method according to any of claims 8-12,

The historical transaction data includes at least one of the following:

The number of historical transactions, the historical transaction amount, and the number of transaction recipients involved in historical transactions.
A method according to any of claims 8-12,

The behavior data includes at least one of the following:

The transaction recipient's information, browsing duration, and browsed resource information browsed by the transaction payer, and the transaction recipient's information includes at least the transaction recipient's credit value, product category, commodity value distribution, and health level.
A device for determining the transfer of a false resource, comprising:

An acquiring unit, acquiring historical resource transfer data of the resource transfer party within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transfer party within a second predetermined time period before the resource to be verified is transferred ;

The determining unit determines, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

The fake resource transfer model is trained based on historical resource transfer training data and corresponding behavior training data.
A device for determining a fraudulent transaction, comprising:

Obtaining a unit, acquiring historical transaction data of a transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

a determining unit, determining, according to the historical resource transfer data, the behavior data, and a fraudulent transaction model, whether the transaction to be verified is a fraudulent transaction;

The false transaction model is trained based on historical transaction training data and corresponding behavioral training data.
An electronic device comprising:

Processor;

A memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:

Obtaining, by the resource transferee, historical resource transfer data within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transferee within a second predetermined time period before the resource to be verified is transferred;

Determining, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

The fake resource transfer model is trained based on historical resource transfer training data and corresponding behavior training data.
A computer readable storage medium storing one or more programs, the one or more programs, when executed by an electronic device including a plurality of applications, causing the electronic device to perform the following operations :

Obtaining, by the resource transferee, historical resource transfer data within a first predetermined time period before the resource to be verified is transferred, and behavior data of the resource transferee within a second predetermined time period before the resource to be verified is transferred;

Determining, according to the historical resource transfer data, the behavior data, and the fake resource transfer model, whether the resource transfer to be verified is a false resource transfer;

The fake resource transfer model is trained based on historical resource transfer training data and corresponding behavior training data.
An electronic device comprising:

Processor;

A memory arranged to store computer executable instructions that, when executed, cause the processor to perform the following operations:

Obtaining historical transaction data of the transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical resource transfer data, the behavior data, and a fraudulent transaction model;

The false transaction model is trained based on historical transaction training data and corresponding behavioral training data.
A computer readable storage medium storing one or more programs, the one or more programs, when executed by an electronic device including a plurality of applications, causing the electronic device to perform the following operations :

Obtaining historical transaction data of the transaction payer within a first predetermined time period before the transaction to be verified, and behavior data of the transaction payer within a second predetermined time period before the transaction to be verified;

Determining whether the transaction to be verified is a fraudulent transaction based on the historical resource transfer data, the behavior data, and a fraudulent transaction model;

The false transaction model is trained based on historical transaction training data and corresponding behavioral training data.