Detailed Description
For the purposes, technical solutions and advantages of the present application, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.
In order to solve the problem that the existing risk identification scheme needs to train corresponding risk identification models for different risk scenes respectively, so that the waste of computing resources is caused, and meanwhile, the deployment and maintenance costs of the risk identification models are increased, the embodiment of the specification provides a method for determining the risk identification scheme.
Specifically, a schematic implementation flow diagram of a method for determining a risk identification scheme provided in one or more embodiments of the present disclosure is shown in fig. 1, where the method of fig. 1 may be performed by a device for determining a risk identification scheme, or in other words, the method of fig. 1 may be performed by a computer, a server, or a device, and includes:
step 110, performing standardization processing on the risk related data of the risk scene to be applied according to a preset standardization processing rule to obtain the risk related data after the standardization processing.
It should be understood that, in order to improve the processing efficiency of the risk related data of the risk scene to be applied, the embodiment of the present disclosure may perform standardized processing on the risk related data of the risk scene to be applied according to a preset standardized processing rule. Specifically, key fields in risk related data of a risk scenario to be applied may be extracted, and filled in corresponding data fields according to a preset standardized format.
For example, a standardized data sequence may be pre-established, where the standardized data sequence may include data such as a transaction serial number, a device number, a buyer ID, a moment when the buyer opens a commodity page, a transaction creation moment, a seller ID, longitude and latitude coordinates of a geographic area where the buyer and the seller are located, and a standardized format corresponding to the data. After acquiring the risk related data of the risk scenario to be applied, the data such as the transaction serial number, the equipment number, the buyer ID, the transaction creation time, the seller ID, the longitude and latitude coordinates of the geographic area where the buyer and the seller are located in the risk related data of the risk scenario can be extracted and filled in the corresponding position according to the corresponding standardized format.
Step 120, determining target risk index calculation logic corresponding to the normalized risk related data from a preset risk index calculation logic library.
Optionally, the target risk indicator calculation logic is configured to determine a risk indicator of a risk scenario to be applied based on the normalized risk related data.
It should be appreciated that the risk situation faced by a risk scenario may be measured by a corresponding risk indicator, which may be empirically set. For example, in the above-mentioned online transaction scenario, the length of the transaction period from browsing the commodity to completing the commodity can be used as a risk indicator for measuring whether the transaction is a false transaction. In practical application, the risk index can be calculated based on original risk related data in a risk scene through a plurality of preset formulas, and can also be directly extracted from the standardized risk related data. Such as transaction time period, the risk indicator may be represented by a preset formula: transaction period = transaction creation time-merchandise page opening time.
For another example, in identifying a false transaction scenario, whether a social relationship exists between the buyer and the seller may also be used as a risk indicator for measuring whether the transaction between the buyer and the seller is a false transaction. In particular, whether a social relationship exists between the buyer and the seller may be determined based on whether there is some correlation between the identity information of the buyer and the seller (e.g., native, graduation, work unit, etc. information). In addition, the difference between the time length of the transaction of the buyer and the time length of the last transaction can also be used as a risk indicator for measuring whether the transaction between the buyer and the seller in the transaction is a false transaction. Specifically, the following formulas may be used, respectively: the time difference of two consecutive transactions = (this transaction creation time-commodity page opening time in this transaction) - (last transaction creation time-commodity page opening time in last transaction) is determined.
Optionally, since some redundant or abnormal data are often included in the risk related data of the risk scenario to be applied, in order to avoid increasing the calculation pressure in the risk indicator calculation process and the model training process for this part of data, the embodiment of the present disclosure may perform a data cleaning operation on the risk related data after the normalization process, so as to remove the redundant or abnormal data in the risk related data after the normalization process. Then, after performing standardization processing on the risk related data of the risk scene to be applied according to a preset standardization format to obtain the risk related data after the standardization processing, the method further includes:
and performing data cleaning operation on the standardized risk related data according to a preset data cleaning rule so as to remove abnormal data in the standardized risk related data.
It should be appreciated that data cleansing is typically a process of re-examining and checking data in order to remove duplicates, correct errors that exist, and provide data consistency. In order to improve the efficiency and accuracy of data cleansing, the embodiment of the present disclosure may preset a data cleansing rule corresponding to risk-related data (i.e., the preset data cleansing rule described above) based on the characteristics of the risk-related data.
Step 130, determining training and recognition logic of the target risk recognition model corresponding to the normalized risk related data from the training and recognition logic library of the risk recognition model.
Optionally, the training and identifying logic of the target risk identification model is configured to train to obtain a target risk identification model for risk assessment of the risk scenario to be applied based on the risk index of the risk scenario to be applied.
It should be appreciated that, to increase the efficiency of determining a risk identification scheme, embodiments of the present disclosure may pre-configure a training and identification logic library of risk identification models, which may include training and identification logic for some trained risk identification models and some risk identification models. The trained risk recognition model can be directly called without retraining, and the training and recognition logic of the risk recognition model can be trained to obtain the corresponding risk recognition model based on the standardized risk indexes.
It should be understood that if the training and recognition logic of the target risk recognition model corresponding to the normalized risk related data is determined from the training and recognition logic library of the risk recognition model, the corresponding risk recognition model may be obtained by training based on the risk indexes of the multiple sets of risk scenes to be applied, and the recognition threshold of the risk recognition model may be determined by the recognition logic of the target risk recognition model. The risk indexes of the multiple groups of risk scenes to be applied are determined based on the standardized risk related data and the target risk index calculation logic in the multiple groups of risk scenes to be applied.
Step 140, determining a risk identification scheme for evaluating the risk scenario to be applied based on the target risk indicator calculation logic and the training and identification logic of the target risk identification model.
Optionally, in order to facilitate deployment and implementation of the risk recognition scheme, in this embodiment of the present disclosure, the target risk indicator calculation logic and the training and recognition logic of the target risk recognition model may be packaged into one component, that is, the target risk recognition component, so that when an application party deploys and implements the risk recognition scheme, the application party does not need to know the internal risk recognition logic, and only needs to input the normalized risk related data into the target risk recognition component, and then, after determining the risk recognition scheme for evaluating the risk scenario to be applied based on the target risk indicator calculation logic and the training and recognition logic of the target risk recognition model, the method further includes:
and packaging the target risk index calculation logic and the training and recognition logic of the target risk recognition model to obtain a packaged target risk recognition component, wherein the target risk recognition component is used for evaluating a risk scene to be applied.
It should be understood that, in practical application, the risk related data after the normalization processing should first obtain a risk indicator based on the target risk indicator calculation logic, and then use the obtained risk indicator as the training of the target risk recognition model and the input of the recognition logic to output the corresponding risk recognition result.
Then, after encapsulating the target risk indicator calculation logic and the training and recognition logic of the target risk recognition model to obtain an encapsulated target risk recognition component, the method further includes:
and taking the standardized risk related data as the input of the target risk identification component to output a risk assessment result of a risk scene to be applied.
Optionally, in order to provide convenience for an application party of the risk identification scheme, so that the application party of the risk identification scheme does not need to know its content processing logic when deploying and implementing the risk identification scheme, the embodiment of the specification may package the preset standardized processing rule, the preset data cleansing rule, the target risk index calculation logic and the training and identifying logic of the target risk identification model into one component, namely, the target risk identification component, so that the application party only needs to input the risk related data into the target risk identification component, then, in order to determine the risk identification scheme for evaluating the risk scene to be applied based on the target risk index calculation logic and the training and identifying logic of the target risk identification model, the method includes:
and packaging a preset standardized processing rule, a preset data cleaning rule, a target risk index calculation logic and a training and identifying logic of a target risk identification model to obtain a packaged target risk identification component, wherein the target risk identification component is used for evaluating a risk scene to be applied.
As shown in fig. 2, the risk recognition scheme provided in the embodiment of the present disclosure is packaged as a schematic diagram of the target risk recognition component 20, in fig. 2, namely, a node 21 (preset standardized processing rule), a node 22 (preset data cleansing rule), a node 23 (target risk index calculation logic), a node 24 (training logic of a target risk recognition model), a node 25 (recognition logic of a target risk recognition model), and a node 26 (risk result output) are packaged as one component, namely, the target risk recognition component 20. The target risk index calculation logic is determined based on risk related data of a risk scene to be applied and a preset risk index calculation logic library, the training logic of the target risk identification model is determined based on risk related data of the risk scene to be applied and a preset risk identification model training logic library, and the identification logic of the target risk identification model is determined based on risk related data of the risk scene to be applied and a preset risk identification model identification logic library.
Then, after the preset standardized processing rule, the preset data cleaning rule, the target risk index calculation logic and the training and identifying logic of the target risk identification model are packaged to obtain the packaged target risk identification component, the method further includes:
and taking the risk related data of the risk scene to be applied as the input of the target risk identification component to output a risk assessment result of the risk scene to be applied.
When determining a risk identification scheme for a risk scene to be applied, carrying out standardized processing on risk related data of the risk scene to be applied according to a preset standardized processing rule to obtain standardized processed risk related data; determining target risk index calculation logic corresponding to the standardized risk related data from a preset risk index calculation logic library; determining training and identifying logic of a target risk identification model corresponding to the standardized risk related data from a training and identifying logic library of the risk identification model; finally, a risk recognition scheme for evaluating the risk scenario to be applied can be determined based on the target risk indicator calculation logic and the training and recognition logic of the target risk recognition model. Because the training and identifying logic of the risk index calculation logic and the risk identifying model are pre-configured, when facing complex scenes comprising a plurality of risk scenes, the training and identifying logic of the configured risk index calculation logic and the risk identifying model can be directly called, and operations such as model training and the like do not need to be performed in a large amount of time, so that the determining efficiency of a risk identifying scheme is improved.
Fig. 3 is a schematic structural diagram of a determining apparatus 300 of a risk identification scheme provided in the embodiment of the present disclosure. Referring to fig. 3, in a software embodiment, a determining apparatus 300 of a risk identification scheme may include a processing unit 301, a first determining unit 302, a second determining unit 303, and a third determining unit 302, wherein:
the processing unit 301 performs standardization processing on risk related data of a risk scene to be applied according to a preset standardization processing rule to obtain risk related data after the standardization processing;
a first determining unit 302, configured to determine target risk indicator calculation logic corresponding to the normalized risk related data from a preset risk indicator calculation logic library;
a second determining unit 303 that determines training and recognition logic of a target risk recognition model corresponding to the normalized risk-related data from a training and recognition logic library of risk recognition models;
the third determining unit 304 determines a risk identification scheme for evaluating the risk scenario to be applied based on the target risk indicator calculation logic and the training and identification logic of the target risk identification model.
When determining a risk identification scheme for a risk scene to be applied, the processing unit 301 is capable of performing standardized processing on risk related data of the risk scene to be applied according to a preset standardized processing rule to obtain standardized risk related data; determining target risk index calculation logic corresponding to the normalized risk related data from a preset risk index calculation logic library through a first determination unit 302; and can determine the training and recognition logic of the target risk recognition model corresponding to the normalized risk-related data from the training and recognition logic library of risk recognition models by the second determination unit 303; finally, by means of the third determination unit 304, a risk recognition scheme for evaluating the risk scenario to be applied can be determined on the basis of the target risk indicator calculation logic and the training and recognition logic of the target risk recognition model. Because the training and identifying logic of the risk index calculation logic and the risk identifying model are pre-configured, when facing complex scenes comprising a plurality of risk scenes, the training and identifying logic of the configured risk index calculation logic and the risk identifying model can be directly called, and operations such as model training and the like do not need to be performed in a large amount of time, so that the determining efficiency of a risk identifying scheme is improved.
Optionally, in an embodiment, the target risk indicator calculation logic is configured to determine a risk indicator of the risk scenario to be applied based on the normalized risk related data.
Optionally, in an embodiment, the training and identifying logic of the target risk identification model is configured to train to obtain a target risk identification model for risk assessment of the risk scenario to be applied based on the normalized risk indicator.
Optionally, in an embodiment, after the third determining unit 304 determines a risk identification scheme for evaluating the risk scenario to be applied based on the target risk indicator calculation logic and the training and identification logic of the target risk identification model, the apparatus further comprises:
the first packaging unit 305 packages the target risk indicator calculation logic and the training and identifying logic of the target risk identification model, to obtain a packaged target risk identification component, where the target risk identification component is used to evaluate the risk scenario to be applied.
Optionally, in an embodiment, after the processing unit 301 performs normalization processing on the risk related data of the risk scenario to be applied according to a preset normalization format to obtain risk related data after the normalization processing, the apparatus further includes:
and a data cleaning unit 306, configured to perform a data cleaning operation on the normalized risk related data according to a preset data cleaning rule, so as to remove abnormal data in the normalized risk related data.
Optionally, in an embodiment, after the third determining unit 304 determines a risk identification scheme for evaluating the risk scenario to be applied based on the target risk indicator calculation logic and the training and identification logic of the target risk identification model, the apparatus further comprises:
and a second packaging unit 307, configured to package the preset standardized processing rule, the preset data cleaning rule, the target risk indicator calculation logic, and the training and identifying logic of the target risk identification model, to obtain a packaged target risk identification component, where the target risk identification component is used to evaluate the risk scenario to be applied.
Optionally, in an embodiment, after the first packaging unit 305 packages the target risk indicator calculation logic and the training and identifying logic of the target risk identification model, the apparatus further includes:
the first evaluation unit 308 takes the risk related data after the normalization processing as an input of the target risk identification component to output a risk evaluation result of a risk scene to be applied.
Optionally, in an embodiment, after the second packaging unit 307 packages the preset standardized processing rule, the preset data cleansing rule, the target risk indicator calculation logic, and the training and identifying logic of the target risk identification model, the apparatus further includes:
the second evaluation unit 309 takes the risk related data of the risk scenario to be applied as the input of the target risk identification component, so as to output a risk evaluation result of the risk scenario to be applied.
The determining device 300 of the risk identification scheme can implement the method of the method embodiment of fig. 1 to 2, and specifically, the determining method of the risk identification scheme of the embodiment shown in fig. 1 to 2 may be referred to, and will not be described again.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. Referring to fig. 4, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.
The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 4, but not only one bus or type of bus.
And the memory is used for storing programs. In particular, the program may include program code including computer-operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.
The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to form the determining device of the risk identification scheme on the logic level. The processor is used for executing the programs stored in the memory and is specifically used for executing the following operations:
carrying out standardized processing on risk related data of a risk scene to be applied according to a preset standardized processing rule to obtain risk related data after the standardized processing;
determining target risk index calculation logic corresponding to the standardized risk related data from a preset risk index calculation logic library;
determining training and identifying logic of a target risk identification model corresponding to the standardized risk related data from a training and identifying logic library of the risk identification model;
and determining a risk identification scheme for evaluating the risk scene to be applied based on the target risk index calculation logic and the training and identification logic of the target risk identification model.
The method for determining the risk identification scheme disclosed in the embodiment shown in fig. 1-2 of the present specification can be applied to a processor or implemented by the processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The various methods, steps, and logic blocks disclosed in one or more embodiments of the present description may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with one or more embodiments of the present disclosure may be embodied directly in a hardware decoding processor or in a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.
The electronic device may further execute the method for determining the risk identification scheme of fig. 1-2, which is not described herein.
Of course, other implementations, such as a logic device or a combination of hardware and software, are not excluded from the electronic device of the present application, that is, the execution subject of the following processing flow is not limited to each logic unit, but may be hardware or a logic device.
The present embodiments also provide a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a portable electronic device comprising a plurality of application programs, enable the portable electronic device to perform the method of the embodiment of fig. 1, and in particular to:
carrying out standardized processing on risk related data of a risk scene to be applied according to a preset standardized processing rule to obtain risk related data after the standardized processing;
determining target risk index calculation logic corresponding to the standardized risk related data from a preset risk index calculation logic library;
determining training and identifying logic of a target risk identification model corresponding to the standardized risk related data from a training and identifying logic library of the risk identification model;
and determining a risk identification scheme for evaluating the risk scene to be applied based on the target risk index calculation logic and the training and identification logic of the target risk identification model.
Of course, in addition to the software implementation, the electronic device in this specification does not exclude other implementations, such as a logic device or a combination of software and hardware, that is, the execution subject of the following process is not limited to each logic unit, but may also be hardware or a logic device.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In summary, the foregoing description is only a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present disclosure, is intended to be included within the scope of one or more embodiments of the present disclosure.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, 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, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer 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 disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.