Summary of the invention
In view of the above problems, it proposes on the present invention overcomes the above problem or at least be partially solved in order to provide one kind
State appraisal procedure, device, electronic equipment and the computer readable storage medium of the client intermediary risk of problem.
According to one aspect of the present invention, a kind of appraisal procedure of client intermediary risk is provided, comprising:
Part client group case correlated characteristic data collection is extracted, the data set may include part client's various dimensions characteristic
According to;
Group, intermediary case data model is constructed based on the complete part client group case correlated characteristic data collection;
Client characteristics data are obtained, the dimension and/or type of the characteristic correspond to group, the intermediary case data model
Input;
Comprehensive score is carried out to the client characteristics data using group, the intermediary case data model, is commented based on the synthesis
Divide intermediary's risk of assessment client.
Optionally, intermediary's risk based on comprehensive score assessment client, further comprises: determining group, intermediary case
Score threshold assesses intermediary's risk according to the comparison result between the score threshold and the comprehensive score.
Optionally, it is described based on the comprehensive score assessment client intermediary's risk, further comprise: according to the data
The client of collection, which scores, to be distributed, and determines group, the intermediary case score threshold.
Optionally, intermediary's risk based on comprehensive score assessment client, further comprises: determining comprehensive score
There are intermediary's risks by client in group, intermediary case score threshold.
Optionally, further includes: to comprehensive score, the client outside group, intermediary case score threshold carries out normal review;And/or
To comprehensive score, the client in group, intermediary case score threshold turns manual examination and verification.
Optionally, the client to comprehensive score in group, intermediary case score threshold carries out user's control.
Optionally, described that group, intermediary case data model is constructed based on the complete part client group case correlated characteristic data collection, into
One step includes: to carry out models fitting using logistic regression algorithm, until obtaining the data model met the requirements.
Optionally, described case correlated characteristic data concentrate data include: location information, related information, network environment,
One of application node information or a variety of combinations.
According to another aspect of the invention, a kind of assessment device of client intermediary risk is provided, comprising:
Data set extraction module is suitable for having extracted part client group case correlated characteristic data collection, and the data set may be adapted to
Part client's various dimensions characteristic;
Model construction module is suitable for constructing group, intermediary case data mould based on the complete part client group case correlated characteristic data collection
Type;
Client characteristics data acquisition module is suitable for obtaining client characteristics data, the dimension and/or type of the characteristic
The input of corresponding group, the intermediary case data model;
Intermediary's risk evaluation module is suitable for carrying out the client characteristics data using group, the intermediary case data model comprehensive
Scoring is closed, intermediary's risk based on comprehensive score assessment client.
Optionally, intermediary's risk evaluation module, is further adapted for: group, intermediary case score threshold is determined, according to described
Comparison result between score threshold and the comprehensive score assesses intermediary's risk.
Optionally, intermediary's risk evaluation module, is further adapted for: it is scored and is distributed according to the client of the data set,
Determine group, the intermediary case score threshold.
Optionally, intermediary's risk evaluation module, is further adapted for: determining comprehensive score in group, intermediary case score threshold
There are intermediary's risks by interior client.
Optionally, further include auditing module, be suitable for: to comprehensive score, the client outside group, intermediary case score threshold is carried out just
Often audit;And/or manual examination and verification are turned to client of the comprehensive score in group, intermediary case score threshold.
Optionally, the auditing module, is further adapted for: to client of the comprehensive score in group, intermediary case score threshold into
Row user's control.
Optionally, the model construction module, is further adapted for: models fitting is carried out using logistic regression algorithm, until
Obtain the data model met the requirements.
Optionally, described case correlated characteristic data concentrate data be suitable for: location information, related information, network environment,
One of application node information or a variety of combinations.
According to another aspect of the invention, a kind of electronic equipment is provided, wherein the electronic equipment includes:
Processor;And
It is arranged to the memory of storage computer executable instructions, executable instruction executes processor when executed
Above-mentioned method.
According to another aspect of the invention, a kind of computer readable storage medium is provided, wherein computer-readable to deposit
Storage media stores one or more programs, and one or more programs when being executed by a processor, realize above-mentioned method.
The utility model has the advantages that
The present invention establishes intermediary's Warning System, comments using machine learning method according to group's case correlated characteristic of client
Estimate intermediary's risk of client, can not only back feeding empirical rule, improve ease for use, while the rate of manslaughtering and operation cost can be reduced,
Intermediary's recognition accuracy and treatment effeciency is greatly improved in the attack for effectively having intercepted group, intermediary case client.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
Fig. 1 shows the flow diagram of the appraisal procedure of client intermediary according to an embodiment of the invention risk.Such as
Shown in Fig. 1, the appraisal procedure of the client intermediary risk of the embodiment of the present invention includes:
S11: having extracted part client group case correlated characteristic data collection, and the data set may include part client's various dimensions feature
Data;
It should be noted that the characteristic of the complete part client carries out conclusion extraction from historical data base, specially
Location information, related information, network environment, application node information of application etc..By from historical data base depth excavate it is more
The various dimensions characteristic of a historic customer, and summarized and the complete part client group case correlated characteristic data collection can be obtained.
S12: group, intermediary case data model is constructed based on the complete part client group case correlated characteristic data collection;
It should be noted that using the characteristic that complete part client group case correlated characteristic data are concentrated as machine learning model
Training sample, the model that can have portrayed part client group case characteristic is trained in the way of machine learning, thus go out building in
Be situated between group's case data model, and the output of model is comprehensive score.
S13: obtaining client characteristics data, and the dimension and/or type of the characteristic correspond to group, the intermediary case data
The input of model;
It should be noted that system obtains the feature of current loaning bill client automatically during client carries out debit operation
Data, the dimension and type of this feature data correspond to the input of group, the intermediary case data model.
S14: comprehensive score is carried out to the client characteristics data using group, the intermediary case data model, based on described comprehensive
Close intermediary's risk of scoring assessment client.
It should be noted that recording the application of this user when online application user starts to apply in platform
Then these data are input to group, intermediary case number by the data such as location information, related information, network environment, application node information
According in model, the data image of this user is portrayed using group, intermediary case data model, exports the comprehensive score corresponding to it, system
Judge whether client is intermediary further according to comprehensive score, to realize intermediary's risk identification.
The embodiment of the present invention establishes intermediary's Warning System according to the related information of client using machine learning method,
Assess client intermediary's risk, can not only back feeding empirical rule, improve ease for use, while can reduce the rate of manslaughtering and operation at
This, has effectively intercepted the attack of group, intermediary case client, intermediary's recognition accuracy and treatment effeciency is greatly improved.
In a kind of optional embodiment of the embodiment of the present invention, based on described described in the S14 in method shown in Fig. 1
Comprehensive score assess client intermediary's risk, further comprise: determining group, intermediary case score threshold, according to the score threshold with
Comparison result between the comprehensive score assesses intermediary's risk.
Wherein intermediary's risk based on comprehensive score assessment client further comprises: according to the data set
Client score distribution, determine group, the intermediary case score threshold.Specifically, system is scored by the client of analyzing and training sample
Distribution, i.e., the client of the group's case correlated characteristic extracted from historical data base, which scores, to be distributed, and then therefrom finds group, intermediary case visitor
The scoring of client corresponding to family, is scored based on the client and determines group, intermediary case score threshold.
Intermediary's risk based on comprehensive score assessment client, further comprises: determining comprehensive score in intermediary
There are intermediary's risks by client in group's case score threshold, specifically, for the client in threshold value, it is believed that there are intermediary's risks and right
The application of the client carries out turning artificial treatment, determines whether the client is case member, intermediary or an intermediary by manual examination and verification
It acts on behalf, if it is not, then entering normal review process by processing;If so, refusal is handled.Client outside for threshold value enters
Normal review process, finally obtains auditing result.
Further, the client to comprehensive score in group, intermediary case score threshold carries out user's control.
In a kind of optional embodiment of the embodiment of the present invention, based on described described in the S12 in method shown in Fig. 1
Complete part client group case correlated characteristic data collection constructs group, intermediary case data model, further comprises: being calculated using logistic regression (LR)
Method carries out models fitting, until obtaining the data model met the requirements.
In a kind of optional embodiment of the embodiment of the present invention, part has been extracted described in the S11 in method shown in Fig. 1
Client group's case correlated characteristic data collection further comprises: first carrying out text to the characteristic that depth is excavated from database
The pretreatment such as analysis, normalization, branch mailbox has just created part client group case phase with it after denoising to realize to data de-noising
Close characteristic data set.
The embodiment of the present invention, which is detached from, determines the relevant dependence of intermediary to artificial, the recognition methods of more system is formed, by building
Vertical intermediary's Warning System, assesses intermediary's risk of client, to provide method for early warning, indirect labor determines group, intermediary case wind
Danger, so that the more acurrate more efficiency of manual examination and verification.
Fig. 2 shows the structural schematic diagrams of the assessment device of client intermediary according to an embodiment of the invention risk.Such as
Shown in Fig. 2, the device of the embodiment of the present invention includes:
Data set extraction module 21 is suitable for having extracted part client group case correlated characteristic data collection, and the data set may be adapted to
The characteristic of complete part client various dimensions characteristic, the complete part client carries out conclusion extraction from historical data base, specifically
For the location information of application, related information, network environment, application node information etc..By from historical data base depth excavate
The various dimensions characteristic of multiple historic customers, and summarized and the complete part client group case correlated characteristic data can be obtained
Collection;
Model construction module 22 is suitable for constructing group, intermediary case data based on the complete part client group case correlated characteristic data collection
Model, specifically, using the characteristic that complete part client group case correlated characteristic data are concentrated as the training sample of machine learning model
This, trains the model that can have portrayed part client group case characteristic in the way of machine learning, to go out building group, intermediary case number
According to model, the output of model is comprehensive score;
Client characteristics data acquisition module 23 is suitable for obtaining client characteristics data, the dimension and/or class of the characteristic
Type corresponds to the input of group, the intermediary case data model, and specifically, during client carries out debit operation, system obtains automatically
The characteristic of current loaning bill client, the dimension and type of this feature data correspond to the input of group, the intermediary case data model;
Intermediary's risk evaluation module 24 is suitable for carrying out the client characteristics data using group, the intermediary case data model
Comprehensive score, based on intermediary's risk of comprehensive score assessment client, specifically, when online application user starts in platform
When being applied, the data such as location information, related information, network environment, the application node information of the application of this user are recorded, so
These data are input in group, intermediary case data model afterwards, the data shape of this user is portrayed using group, intermediary case data model
As exporting the comprehensive score corresponding to it, system judges whether client is intermediary further according to comprehensive score, to realize
Intermediary's risk identification.
The embodiment of the present invention establishes intermediary's Warning System according to the related information of client using machine learning method,
Assess client intermediary's risk, can not only back feeding empirical rule, improve ease for use, while can reduce the rate of manslaughtering and operation at
This, has effectively intercepted the attack of group, intermediary case client, intermediary's recognition accuracy and treatment effeciency is greatly improved.
In another embodiment of the present invention, intermediary's risk evaluation module 24 of Fig. 2 shown device, is further adapted for:
It determines group, intermediary case score threshold, the intermediary is assessed according to the comparison result between the score threshold and the comprehensive score
Risk.
Wherein intermediary's risk based on comprehensive score assessment client further comprises: according to the data set
Client score distribution, determine group, the intermediary case score threshold.Specifically, system is scored by the client of analyzing and training sample
Distribution, i.e., the client of the group's case correlated characteristic extracted from historical data base, which scores, to be distributed, and then therefrom finds group, intermediary case visitor
The scoring of client corresponding to family, is scored based on the client and determines group, intermediary case score threshold
Intermediary's risk based on comprehensive score assessment client, further comprises: determining comprehensive score in intermediary
There are intermediary's risks by client in group's case score threshold, specifically, for the client in threshold value, it is believed that there are intermediary's risks and right
The application of the client carries out turning artificial treatment, determines whether the client is case member, intermediary or an intermediary by manual examination and verification
It acts on behalf, if it is not, then entering normal review process by processing;If so, refusal is handled.Client outside for threshold value enters
Normal review process, finally obtains auditing result.
Further, the client to comprehensive score in group, intermediary case score threshold carries out user's control.
In another embodiment of the present invention, the model construction module 22 of Fig. 2 shown device, is further adapted for: using
Logistic regression (LR) algorithm carries out models fitting, until obtaining the data model met the requirements.
In another embodiment of the present invention, the data set extraction module 21 of Fig. 2 shown device, is further adapted for: first
The pretreatment such as text analyzing, normalization, branch mailbox is carried out to the characteristic that depth is excavated from database, to realize to data
Denoising has just created part client group case correlated characteristic data collection with it after denoising.
The device of the embodiment of the present invention can be used for executing above method embodiment, and principle is similar with technical effect, this
Place repeats no more.
It should be understood that
Algorithm and display be not inherently related to any certain computer, virtual bench or other equipment provided herein.
Various fexible units can also be used together with teachings based herein.As described above, it constructs required by this kind of device
Structure be obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use various
Programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e. required to protect
Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
Meaning one of can in any combination mode come using.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
Microprocessor or digital signal processor (DSP) realize the wearing state of detection electronic equipment according to an embodiment of the present invention
Device in some or all components some or all functions.The present invention is also implemented as executing institute here
Some or all device or device programs of the method for description are (for example, computer program and computer program produce
Product).It is such to realize that program of the invention can store on a computer-readable medium, or can have one or more
The form of signal.Such signal can be downloaded from an internet website to obtain, and perhaps be provided on the carrier signal or to appoint
What other forms provides.
For example, Fig. 3 shows the structural schematic diagram of electronic equipment according to an embodiment of the invention.The electronic equipment passes
It include processor 31 and the memory 32 for being arranged to storage computer executable instructions (program code) on system.Memory 32 can
To be the Electronic saving of such as flash memory, EEPROM (electrically erasable programmable read-only memory), EPROM, hard disk or ROM etc
Device.Memory 32 has the program code 34 stored for executing any method and step in shown in FIG. 1 and each embodiment
Memory space 33.For example, the memory space 33 for program code may include being respectively used to realize in above method
Each program code 34 of various steps.These program codes can be read from one or more computer program product or
Person is written in this one or more computer program product.These computer program products include such as hard disk, compact-disc
(CD), the program code carrier of storage card or floppy disk etc.Such computer program product is usually described in such as Fig. 4
Computer readable storage medium.The computer readable storage medium can have 32 class of memory in the electronic equipment with Fig. 3
Like memory paragraph, the memory space etc. of arrangement.Program code can for example be compressed in a suitable form.In general, storage unit is deposited
Contain the program code 41 for executing steps of a method in accordance with the invention, it can read by such as 31 etc processor
Program code causes the electronic equipment to execute in method described above when these program codes are run by electronic equipment
Each step.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.