CN107767259A - Loan risk control method, electronic installation and readable storage medium storing program for executing - Google Patents
Loan risk control method, electronic installation and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN107767259A CN107767259A CN201710916484.8A CN201710916484A CN107767259A CN 107767259 A CN107767259 A CN 107767259A CN 201710916484 A CN201710916484 A CN 201710916484A CN 107767259 A CN107767259 A CN 107767259A
- Authority
- CN
- China
- Prior art keywords
- user
- credit risk
- application program
- default
- grade
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Technology Law (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention relates to a kind of loan risk control method, electronic installation and readable storage medium storing program for executing, this method includes：The seed user of default different credit risk grades is associated with the user's portrait pre-established, obtains user's representation data of different credit risk grades；User's representation data based on different credit risk grades trains to obtain the hierarchy model of credit risk grade, and is classified using the hierarchy model to presetting the user in application program, is divided into different credit risk grades；The different credit risk grades according to corresponding to user in default application program, and risk score is carried out to the user preset in application program by default code of points, the user that preset requirement is only reached to scoring shows default loan entrance.Present invention reduces the risk of finance company, the conversion ratio of lifting loan user.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of loan risk control method, electronic installation and readable
Storage medium.
Background technology
Current each finance company mainly by the way of extensively casting net, that is, is directed to when promoting loan transaction on water conservancy diversion APP
All users show loan entrance, can also not pass through arbitrarily in view of the risk control problem of user's loan, client inferior
Loan entrance handles loan, increases credit risk, reduces loan success rate.
The content of the invention
It is an object of the invention to provide a kind of loan risk control method, electronic installation and readable storage medium storing program for executing, it is intended to
Reduce credit risk.
To achieve the above object, the present invention provides a kind of electronic installation, and the electronic installation includes memory, processor,
The credit risk control system that can be run on the processor, the credit risk control system are stored with the memory
Following steps are realized during by the computing device：
A, the seed user of default different credit risk grades is associated with the user's portrait pre-established, obtained
User's representation data of different credit risk grades；
B, user's representation data based on different credit risk grades trains to obtain the hierarchy model of credit risk grade, and
It is classified using the hierarchy model to presetting the user in application program, is divided into different credit risk grades；
C, according to different credit risk grades corresponding to user in default application program, and by default code of points to default
User in application program carries out risk score, and the user that preset requirement is only reached to scoring shows default loan entrance.
Preferably, credit risk grade includes four grades a, b, c, d, and the training process of the hierarchy model is as follows：
User's representation data of different credit risk grades is trained using multiclass partitioning, is extracting training set
When extract user's representation data vector corresponding to credit risk grade a respectively as positive collection, credit risk grade b, c, d institutes
Corresponding user's representation data vector is as negative collection, to obtain the first training set；Extract the user corresponding to credit risk grade b
Representation data vector collects as positive, extracts user's representation data vector corresponding to credit risk grade a, c, d and collects as negative, with
Obtain the second training set；User's representation data vector corresponding to credit risk grade c is extracted as positive collection, credit risk grade
A, user's representation data vector corresponding to b, d is as negative collection, to obtain the 3rd training set；It is right to extract credit risk grade d institutes
The user's representation data vector answered is as positive collection, and user's representation data vector corresponding to credit risk grade a, b, c is as negative
Collection, to obtain the 4th training set；This four training sets are trained respectively, four support vector machines are obtained, as loan
The hierarchy model of risk class.
Preferably, it is described to carry out classification to presetting the user in application program using the hierarchy model and include：
, will be default when being classified using the hierarchy model to the credit risk grade for presetting user in application program
Four support vector machines that the user data vector of user is utilized respectively in the hierarchy model in application program are surveyed
Examination, classification function value of the user data vector of user in default application program on each support vector machines is obtained, will be had
There is credit risk grade of the credit risk grade as the user corresponding to the support vector machines of maximum classification function value.
Preferably, the basis presets different credit risk grades corresponding to user in application program, and by default scoring
Rule carries out risk score to the user preset in application program to be included：
Behavioral data of the user in default application program is obtained, according to the behavioral data by default application program
It is different behavior classifications that user, which carries out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is corresponding to credit risk grade
Weight coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
In addition, to achieve the above object, the present invention also provides a kind of loan risk control method, the credit risk control
Method includes：
Step 1: the seed user of default different credit risk grades is closed with the user's portrait pre-established
Connection, obtains user's representation data of different credit risk grades；
Step 2: user's representation data based on different credit risk grades trains to obtain the classification mould of credit risk grade
Type, and be classified using the hierarchy model to presetting the user in application program, it is divided into different credit risk grades；
Step 3: the different credit risk grades according to corresponding to user in default application program, and by default code of points
To preset application program on user carry out risk score, only to scoring reach preset requirement user show preset provide a loan into
Mouthful.
Preferably, credit risk grade includes four grades a, b, c, d, and the training process of the hierarchy model is as follows：
User's representation data of different credit risk grades is trained using multiclass partitioning, is extracting training set
When extract user's representation data vector corresponding to credit risk grade a respectively as positive collection, credit risk grade b, c, d institutes
Corresponding user's representation data vector is as negative collection, to obtain the first training set；Extract the user corresponding to credit risk grade b
Representation data vector collects as positive, extracts user's representation data vector corresponding to credit risk grade a, c, d and collects as negative, with
Obtain the second training set；User's representation data vector corresponding to credit risk grade c is extracted as positive collection, credit risk grade
A, user's representation data vector corresponding to b, d is as negative collection, to obtain the 3rd training set；It is right to extract credit risk grade d institutes
The user's representation data vector answered is as positive collection, and user's representation data vector corresponding to credit risk grade a, b, c is as negative
Collection, to obtain the 4th training set；This four training sets are trained respectively, four support vector machines are obtained, as loan
The hierarchy model of risk class.
Preferably, it is described to carry out classification to presetting the user in application program using the hierarchy model and include：
, will be default when being classified using the hierarchy model to the credit risk grade for presetting user in application program
Four support vector machines that the user data vector of user is utilized respectively in the hierarchy model in application program are surveyed
Examination, classification function value of the user data vector of user in default application program on each support vector machines is obtained, will be had
There is credit risk grade of the credit risk grade as the user corresponding to the support vector machines of maximum classification function value.
Preferably, the basis presets different credit risk grades corresponding to user in application program, and by default scoring
Rule carries out risk score to the user preset in application program to be included：
Behavioral data of the user in default application program is obtained, according to the behavioral data by default application program
It is different behavior classifications that user, which carries out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is corresponding to credit risk grade
Weight coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
Preferably, the basis presets different credit risk grades corresponding to user in application program, and by default scoring
Rule carries out risk score to the user preset in application program to be included：
Location data of the user in default application program is obtained, and according to the location data and the loan wind of the user
User's representation data corresponding to dangerous grade is to preset label corresponding to the user is set, then calculates user in default application program
Risk score M formula is：
M=x3*M3+y4*M4
Wherein, M3 is scoring radix corresponding to default different credit risk grades, and x3 is corresponding to credit risk grade
Weight coefficient；M4 is scoring radix corresponding to default different labels, and y4 is weight coefficient corresponding to label.
Further, to achieve the above object, the present invention also provides a kind of computer-readable recording medium, the computer
Readable storage medium storing program for executing is stored with credit risk control system, and the credit risk control system can be held by least one processor
OK, so that the step of at least one computing device loan risk control method described above.
Loan risk control method, system and readable storage medium storing program for executing proposed by the present invention, by by different credit risks etc.
The seed user of level is associated with user's portrait, obtains user's representation data of different credit risk grades, and based on difference
User's representation data of credit risk grade trains to obtain the hierarchy model of credit risk grade, will be pre- using the hierarchy model
If the user in application program is divided into different credit risk grades；Different credit risk grades carry out risk according to corresponding to user
Scoring, the user that preset requirement is only reached to scoring show default loan entrance.Due to can to preset application program on user
Credit risk scoring is carried out, the loan entrance in default application program is only showed to the satisfactory user that scores, is improved
The loan success rate of loan transaction is promoted in the default application program of water conservancy diversion, avoids the undesirable user visitor i.e. inferior that scores
Family also arbitrarily can handle loan by entrance of providing a loan, and reduce the risk of finance company, the conversion ratio of lifting loan user.
Brief description of the drawings
Fig. 1 is each optional application environment schematic diagram of embodiment one of the present invention；
Fig. 2 is the schematic diagram of the hardware structure of the embodiment of electronic installation one in Fig. 1；
Fig. 3 is the schematic flow sheet of the embodiment of loan risk control method one of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.Based on the embodiment in the present invention, those of ordinary skill in the art are not before creative work is made
The every other embodiment obtained is put, belongs to the scope of protection of the invention.
It should be noted that the description for being related to " first ", " second " etc. in the present invention is only used for describing purpose, and can not
It is interpreted as indicating or implies its relative importance or imply the quantity of the technical characteristic indicated by indicating.Thus, define " the
One ", at least one this feature can be expressed or be implicitly included to the feature of " second ".In addition, the skill between each embodiment
Art scheme can be combined with each other, but must can be implemented as basis with those of ordinary skill in the art, when technical scheme
With reference to occurring conflicting or will be understood that the combination of this technical scheme is not present when can not realize, also not in application claims
Protection domain within.
It is each optional application environment schematic diagram of embodiment one of the present invention refering to Fig. 1.
In the present embodiment, present invention can apply to include but not limited to, electronic installation 1, terminal device 2, network 3
Application environment in.Wherein, electronic installation 1 is a kind of can to carry out numerical value automatically according to the instruction for being previously set or storing
Calculating and/or the equipment of information processing.Electronic installation 1 can be computer, can also be single network server, multiple networks
Server group into the server group either cloud being made up of a large amount of main frames or the webserver based on cloud computing, its medium cloud meter
It is one kind of Distributed Calculation, a super virtual computer being made up of the computer collection of a group loose couplings.
Terminal device 2 include, but not limited to any one can with user by keyboard, mouse, remote control, touch pad or
The modes such as person's voice-operated device carry out the electronic product of man-machine interaction, for example, personal computer, tablet personal computer, smart mobile phone, individual
Digital assistants (Personal Digital Assistant, PDA), game machine, IPTV (Internet
Protocol Television, IPTV), intellectual Wearable etc..
The network 3 can be intranet (Intranet), internet (Internet), global system for mobile communications
(Global S credit risks control stem of Mobile communication, GSM), WCDMA (Wideband
Code Division Multiple Access, WCDMA), 4G networks, 5G networks, bluetooth (Bluetooth), the nothing such as Wi-Fi
Line or cable network.Wherein, the electronic installation 1 is communicated with one or more terminal devices 2 respectively by the network 3
Connection.
It is the schematic diagram of 1 one optional hardware structure of electronic installation in Fig. 1 refering to Fig. 2, in the present embodiment, electronic installation 1
It may include, but be not limited only to, the memory 11, processor 12, network interface 13 of connection can be in communication with each other by system bus.Need
It is noted that Fig. 2 illustrate only the electronic installation 1 with component 11-13, it should be understood that being not required for implementing institute
There is the component shown, the more or less component of the implementation that can be substituted.
Wherein, the memory 11 comprises at least a type of readable storage medium storing program for executing, and the readable storage medium storing program for executing includes
Flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memories etc.), random access storage device (RAM), it is static with
Machine access memory (SRAM), read-only storage (ROM), Electrically Erasable Read Only Memory (EEPROM), it is programmable only
Read memory (PROM), magnetic storage, disk, CD etc..In certain embodiments, the memory 11 can be the electricity
The internal storage unit of sub-device 1, such as the hard disk or internal memory of the electronic installation 1.In further embodiments, the memory
11 can also be the plug-in type hard disk being equipped with the External memory equipment of the electronic installation 1, such as the electronic installation 1, intelligence
Storage card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card)
Deng.Certainly, the memory 11 can also both include the internal storage unit of the electronic installation 1 or be set including its external storage
It is standby.In the present embodiment, the memory 11 is generally used for the operating system and types of applications that storage is installed on the electronic installation 1
Software, such as program code of the credit risk control system 10 etc..In addition, the memory 11 can be also used for temporarily
Store the Various types of data that has exported or will export.
The processor 12 can be in certain embodiments central processing unit (Central Processing Unit,
CPU), controller, microcontroller, microprocessor or other data processing chips.The processor 12 is generally used for controlling the electricity
The overall operation of sub-device 1, such as perform the control and processing related to the terminal device 2 progress data interaction or communication
Deng.In the present embodiment, the processor 12 is used to run the program code stored in the memory 11 or processing data, example
The credit risk control system 10 as described in running.
The network interface 13 may include radio network interface or wired network interface, and the network interface 13 is generally used for
Communication connection is established between the electronic installation 1 and other electronic equipments.In the present embodiment, the network interface 13 is mainly used in
The electronic installation 1 is connected with one or more terminal devices 2 by the network 3, in the electronic installation 1 and one
Data transmission channel and communication connection are established between individual or multiple terminal devices 2.
Credit risk control system 10 includes at least one computer-readable instruction being stored in the memory 11, should
At least one computer-readable instruction can be performed by the processor 12, to realize each embodiment of the application.
Wherein, following steps are realized when above-mentioned credit risk control system 10 is performed by the processor 12：
Step S1, the seed user of default different credit risk grades is closed with the user's portrait pre-established
Connection, obtains user's representation data of different credit risk grades.
In the present embodiment, batch of seeds user is obtained first, and the classification of credit risk grade is carried out to seed user.Example
Such as, a collection of representative, generality seed user can be picked out from customer data base by expert, and rule of thumb to kind
Child user carries out the classification of credit risk grade, is such as divided into excessive risk, risk, low-risk credit risk grade.
The user's portrait pre-established is obtained, and by the seed user for the different credit risk grades selected with pre-establishing
User portrait be associated, to obtain user's representation data of different credit risk grades.Wherein, user's portrait is built upon
A series of targeted customer's model on True Datas, it is according to information such as user's social property, habits and customs and consumer behaviors
And the user model of the labeling taken out.The core work for building user's portrait is to be labeled " " to user, and is marked
Label are by analyzing user profile the highly refined signature identification to come.Pre-set user in the present embodiment is drawn a portrait and can be
Directly invoke the well-established user portrait related to loan personnel or by various data sources (such as loan website
Database, QQ, microblogging, wechat, snowball, east wealth social software etc.) drawn a portrait to establish user, wrapped in user's portrait of foundation
Include various attribute tags, such as the social property label and Financial Attribute label of user, as the age, family status, income, occupation,
Consumption habit, whether there is loan documentation, whether did credit card etc..
The seed user of different credit risk grades is associated with corresponding user's portrait comprising attribute tags, i.e.,
Corresponding each attribute mark during every attribute in seed user data such as age, sex, hobby, income are drawn a portrait with user
Label are matched, and the higher seed user data of matching degree and user's portrait are associated, all category during user is drawn a portrait
Property label assign seed user data associated therewith, the user's representation data formed after association.So so that original to only have
The seed user data of base attribute are also provided with for example various social property labels of various attribute tags and gold in user's portrait
Melt attribute tags etc., in order to subsequently establish model and more accurately user in the water conservancy diversion application program to promoting loan transaction
Carry out the classification of credit risk grade.
Step S2, user's representation data based on different credit risk grades train to obtain the classification mould of credit risk grade
Type, and be classified using the hierarchy model to presetting the user in application program, it is divided into different credit risk grades.
In the present embodiment, user's representation data of the different credit risk grades after association is got, different loans can be based on
User's representation data of money risk class trains to obtain the hierarchy model of credit risk grade.In a kind of optional embodiment
In, one-to-many method (one-versus-rest, abbreviation OVR SVMs) can be used to train to obtain hierarchy model, during training successively
User's representation data that the sample of some classification is some credit risk grade is classified as one kind, other remaining samples are classified as
Another kind of, the sample of such k classification has just constructed k support vector machines, is categorized as having by unknown sample during classification
That class of maximum classification function value.In another optional embodiment, one-to-one method (one-versus- can be also used
One, abbreviation OVO SVMs or pairwise) to train to obtain hierarchy model, its way is in any i.e. two kinds of two classes sample
A SVM is designed between user's representation data of credit risk grade, therefore the sample of k classification just needs to design k (k-1)/2
Individual SVM, when classifying to a unknown sample, last who gets the most votes's classification is the classification of the unknown sample.
It is being multiclass partitioning (one vs rest) to different loans using one-to-many method in a kind of embodiment
When user's representation data of money risk class is trained to obtain hierarchy model, it may be assumed that credit risk grade includes four grades
A, b, c, d, then the training process of the hierarchy model is as follows：
User's representation data vector corresponding to credit risk grade a is extracted respectively when training set is extracted as just
Collection, user's representation data vector corresponding to credit risk grade b, c, d is as negative collection, to obtain the first training set；Extract and borrow
User's representation data vector corresponding to money risk class b extracts the user corresponding to credit risk grade a, c, d as positive collection
Representation data vector is as negative collection, to obtain the second training set；Extract user's representation data corresponding to credit risk grade c to
Amount is as positive collection, and user's representation data vector corresponding to credit risk grade a, b, d is as negative collection, to obtain the 3rd training
Collection；Extract user's representation data vector corresponding to credit risk grade d as positive to collect, corresponding to credit risk grade a, b, c
User's representation data vector as negative collection, to obtain the 4th training set；This four training sets are trained respectively, obtain four
Individual support vector machines, the hierarchy model as credit risk grade.
When being classified using the hierarchy model to the credit risk grade for presetting user in application program, extract
User data (name that such as user fills in when being registered in application program, sex, age, the duty of user in default application program
The basic attribute datas such as industry, or user caused user content interested, consumption habit during using application program
Etc.), and by user data vectorization, obtain corresponding to user data vector.Four be utilized respectively in the hierarchy model
Support vector machines are tested user data vector, obtain the user data vector of user in default application program every
Classification function value on individual support vector machines, user corresponding to the support vector machines with maximum classification function value is drawn
As credit risk grade of the credit risk grade as the user of data vector.I.e. by the first training set, the second training set,
This four training sets of three training sets, the 4th training set are trained respectively, then obtain the training knot of four support vector machines
Fruit file, when follow-up test, the test vector corresponding to is utilized respectively this four training result files and tested, most
Each test has a result afterwards, and final result is maximum in this four values one.
Step S3, the different credit risk grades according to corresponding to user in default application program, and by default code of points
To preset application program on user carry out risk score, only to scoring reach preset requirement user show preset provide a loan into
Mouthful.
In the present embodiment, each user in default application program is divided into corresponding different loans using the hierarchy model
After risk class, risk score can be carried out to each user.For example, preset different credit risk grades include it is high, in
Etc. on the upper side, medium four ranks on the lower side, low, then in each user, high, medium credit risk on the lower side on the upper side, medium can be set
Its risk score of the user of grade is low, then only shows default loan entrance to the low user of risk score.Carrying out risk score
When, can also on the basis of the credit risk grade of user, with reference to pre-set business rule, class of subscriber, user tag,
The combined factors such as the current positioning address of user carry out credit risk scoring to it, and user's exhibition of preset requirement is only reached to scoring
Loan entrance is now preset, the excessive risk user that preset requirement is not reached to scoring does not show default loan entrance then, to prevent height
Risk subscribers handle loan transaction by default loan entrance.When showing default loan entrance, can scored according to device number
The terminal device for reaching the user of preset requirement is shown when this presets Application Program Interface in predeterminated position such as bottom, top etc.
APP interfaces key position shows default loan entrance, is provided a loan with facilitating guiding scoring to reach the user of preset requirement.
Compared with prior art, the present embodiment is carried out by the way that the seed user of different credit risk grades and user are drawn a portrait
Association, obtains user's representation data of different credit risk grades, and based on user's representation data of different credit risk grades
Training obtains the hierarchy model of credit risk grade, and the user in default application program is divided into difference using the hierarchy model
Credit risk grade；The different credit risk grades according to corresponding to user carry out risk score, only reach preset requirement to scoring
User show default loan entrance.Due to credit risk scoring can be carried out to the user preset in application program, only to scoring
Satisfactory user shows the loan entrance in default application program, improves to promote in the default application program of water conservancy diversion and borrows
The loan success rate of money business, the undesirable user client i.e. inferior that avoids scoring arbitrarily can also be handled by entrance of providing a loan
Loan, reduce the risk of finance company, the conversion ratio of lifting loan user.
In an optional embodiment, on the basis of above-mentioned Fig. 2 embodiment, the quilt of credit risk control system 10
When the step S3 is realized in the execution of processor 12, in addition to：
Behavioral data of the user in default application program is obtained, according to the behavioral data by default application program
It is different behavior classifications that user, which carries out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is corresponding to credit risk grade
Weight coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
Behavioral data of the user in default application program can be also considered in the present embodiment, and according to the behavior number
It is different behavior classifications according to the user in default application program is carried out into cluster, for example, can be according to different user in default application
Consumer behavior data in program will be clustered often to consume, consuming once in a while, the classification such as not consuming substantially, or are accustomed to wholesale and are disappeared
Take, be accustomed to the classifications such as small amount consumption.According to practical application need to preset different credit risk grades corresponding to score base
Number M1 and its corresponding weight coefficient x1, and scoring radix M2 and its corresponding weight coefficient corresponding to different behavior classifications
Y1, COMPREHENSIVE CALCULATING obtain the risk score M of user.For example, predeterminable different credit risk grades are for example high, medium on the upper side, medium
Scoring radix is respectively 1,0.8,0.6,0.4 corresponding on the lower side, low, and weight coefficient x1 corresponding to credit risk grade is 0.4.In advance
If different behavior classifications are as often consumed, consumption, substantially not consumer other scoring radix are respectively 0.5,0.7,0.9 once in a while,
Weight coefficient corresponding to behavior classification is 0.6.If a user determines that it is high credit risk grade, and root in default application program
Determined that it is according to the behavioral data user of the user in default application program and consume classification once in a while, then the use can be calculated
The risk score M=0.4*1+0.6*0.7=0.82 at family.The risk that each user in default application program is calculated successively is commented
Point, the risk for the higher explanation loan that scores is higher, therefore, use of the preset requirement as being less than a certain predetermined threshold value is only reached to scoring
Family shows default loan entrance.
In the present embodiment, when calculating the risk score of user, the credit risk grade and use of user can be considered
Behavioral data of the family in default application program, the credit risk of user more accurately can be comprehensively evaluated, more accurately to enter
Row credit risk controls.
In an optional embodiment, the credit risk control system 10 is performed by the processor 12 and realizes the step
During rapid S3, in addition to：
Location data of the user in default application program is obtained, and according to the location data and the loan wind of the user
User's representation data corresponding to dangerous grade is to preset label corresponding to the user is set, then calculates user in default application program
Risk score M formula is：
M=x3*M3+y4*M4
Wherein, M3 is scoring radix corresponding to default different credit risk grades, and x3 is corresponding to credit risk grade
Weight coefficient；M4 is scoring radix corresponding to default different labels, and y4 is weight coefficient corresponding to label.
Location data of the user in default application program can be also considered in the present embodiment, and according to the positioning number
According to and the credit risk grade of the user corresponding to user's representation data be the user set corresponding to preset label.For example, can
It is luxury goods market according to the positioning address often occurred in the location data of user, and user's representation data corresponding to the user
In income attribute tags be booming income crowd, then can be the user set corresponding to preset label be " high consumption crowd "；If
Positioning address is common market, and the income attribute tags in user's representation data corresponding to the user are medium income group,
Can be then that label is preset corresponding to the user is set as " medium consumer groups ", etc..Set in advance according to the needs of practical application
Scoring radix M3 and its corresponding weight coefficient x3 corresponding to fixed different credit risk grades, and scored corresponding to different labels
Radix M4 and its corresponding weight coefficient y4, COMPREHENSIVE CALCULATING obtain the risk score M of user.For example, predeterminable different loan wind
Dangerous grade is for example high, it is medium it is on the upper side, it is medium it is on the lower side, low corresponding to scoring radix be respectively 1,0.8,0.6,0.4, credit risk grade
Corresponding weight coefficient x1 is 0.3.Preset the scoring of different labels such as high consumption crowd, medium consumer groups, low consumption crowd
Radix is respectively 0.3,0.5,0.9, and weight coefficient corresponding to behavior classification is 0.7.If a user determines in default application program
It is high credit risk grade, and determines that the label of the user is according to location data of the user in default application program
" medium consumer groups ", then the risk score M=0.3*1+0.5*0.7=0.65 of the user can be calculated.Calculate successively
The risk score of each user on to default application program, the risk for the higher explanation loan that scores is higher, and therefore, only scoring is reached
To preset requirement, such as user less than a certain predetermined threshold value shows default loan entrance.
In the present embodiment, when calculating the risk score of user, the credit risk grade and use of user can be considered
Location data of the family in default application program, the credit risk of user more accurately can be comprehensively evaluated, more accurately to enter
Row credit risk controls.
As shown in figure 3, Fig. 3 is the schematic flow sheet of the embodiment of loan risk control method one of the present invention, the credit risk
Control method comprises the following steps：
Step S10, the seed user of default different credit risk grades is closed with the user's portrait pre-established
Connection, obtains user's representation data of different credit risk grades.
In the present embodiment, batch of seeds user is obtained first, and the classification of credit risk grade is carried out to seed user.Example
Such as, a collection of representative, generality seed user can be picked out from customer data base by expert, and rule of thumb to kind
Child user carries out the classification of credit risk grade, is such as divided into excessive risk, risk, low-risk credit risk grade.
The user's portrait pre-established is obtained, and by the seed user for the different credit risk grades selected with pre-establishing
User portrait be associated, to obtain user's representation data of different credit risk grades.Wherein, user's portrait is built upon
A series of targeted customer's model on True Datas, it is according to information such as user's social property, habits and customs and consumer behaviors
And the user model of the labeling taken out.The core work for building user's portrait is to be labeled " " to user, and is marked
Label are by analyzing user profile the highly refined signature identification to come.Pre-set user in the present embodiment is drawn a portrait and can be
Directly invoke the well-established user portrait related to loan personnel or by various data sources (such as loan website
Database, QQ, microblogging, wechat, snowball, east wealth social software etc.) drawn a portrait to establish user, wrapped in user's portrait of foundation
Include various attribute tags, such as the social property label and Financial Attribute label of user, as the age, family status, income, occupation,
Consumption habit, whether there is loan documentation, whether did credit card etc..
The seed user of different credit risk grades is associated with corresponding user's portrait comprising attribute tags, i.e.,
Corresponding each attribute mark during every attribute in seed user data such as age, sex, hobby, income are drawn a portrait with user
Label are matched, and the higher seed user data of matching degree and user's portrait are associated, all category during user is drawn a portrait
Property label assign seed user data associated therewith, the user's representation data formed after association.So so that original to only have
The seed user data of base attribute are also provided with for example various social property labels of various attribute tags and gold in user's portrait
Melt attribute tags etc., in order to subsequently establish model and more accurately user in the water conservancy diversion application program to promoting loan transaction
Carry out the classification of credit risk grade.
Step S20, user's representation data based on different credit risk grades train to obtain the classification of credit risk grade
Model, and be classified using the hierarchy model to presetting the user in application program, it is divided into different credit risk grades.
In the present embodiment, user's representation data of the different credit risk grades after association is got, different loans can be based on
User's representation data of money risk class trains to obtain the hierarchy model of credit risk grade.In a kind of optional embodiment
In, one-to-many method (one-versus-rest, abbreviation OVR SVMs) can be used to train to obtain hierarchy model, during training successively
User's representation data that the sample of some classification is some credit risk grade is classified as one kind, other remaining samples are classified as
Another kind of, the sample of such k classification has just constructed k support vector machines, is categorized as having by unknown sample during classification
That class of maximum classification function value.In another optional embodiment, one-to-one method (one-versus- can be also used
One, abbreviation OVO SVMs or pairwise) to train to obtain hierarchy model, its way is in any i.e. two kinds of two classes sample
A SVM is designed between user's representation data of credit risk grade, therefore the sample of k classification just needs to design k (k-1)/2
Individual SVM, when classifying to a unknown sample, last who gets the most votes's classification is the classification of the unknown sample.
It is being multiclass partitioning (one vs rest) to different loans using one-to-many method in a kind of embodiment
When user's representation data of money risk class is trained to obtain hierarchy model, it may be assumed that credit risk grade includes four grades
A, b, c, d, then the training process of the hierarchy model is as follows：
User's representation data vector corresponding to credit risk grade a is extracted respectively when training set is extracted as just
Collection, user's representation data vector corresponding to credit risk grade b, c, d is as negative collection, to obtain the first training set；Extract and borrow
User's representation data vector corresponding to money risk class b extracts the user corresponding to credit risk grade a, c, d as positive collection
Representation data vector is as negative collection, to obtain the second training set；Extract user's representation data corresponding to credit risk grade c to
Amount is as positive collection, and user's representation data vector corresponding to credit risk grade a, b, d is as negative collection, to obtain the 3rd training
Collection；Extract user's representation data vector corresponding to credit risk grade d as positive to collect, corresponding to credit risk grade a, b, c
User's representation data vector as negative collection, to obtain the 4th training set；This four training sets are trained respectively, obtain four
Individual support vector machines, the hierarchy model as credit risk grade.
When being classified using the hierarchy model to the credit risk grade for presetting user in application program, extract
User data (name that such as user fills in when being registered in application program, sex, age, the duty of user in default application program
The basic attribute datas such as industry, or user caused user content interested, consumption habit during using application program
Etc.), and by user data vectorization, obtain corresponding to user data vector.Four be utilized respectively in the hierarchy model
Support vector machines are tested user data vector, obtain the user data vector of user in default application program every
Classification function value on individual support vector machines, user corresponding to the support vector machines with maximum classification function value is drawn
As credit risk grade of the credit risk grade as the user of data vector.I.e. by the first training set, the second training set,
This four training sets of three training sets, the 4th training set are trained respectively, then obtain the training knot of four support vector machines
Fruit file, when follow-up test, the test vector corresponding to is utilized respectively this four training result files and tested, most
Each test has a result afterwards, and final result is maximum in this four values one.
Step S30, the different credit risk grades according to corresponding to user in default application program, and by default code of points
To preset application program on user carry out risk score, only to scoring reach preset requirement user show preset provide a loan into
Mouthful.
In the present embodiment, each user in default application program is divided into corresponding different loans using the hierarchy model
After risk class, risk score can be carried out to each user.For example, preset different credit risk grades include it is high, in
Etc. on the upper side, medium four ranks on the lower side, low, then in each user, high, medium credit risk on the lower side on the upper side, medium can be set
Its risk score of the user of grade is low, then only shows default loan entrance to the low user of risk score.Carrying out risk score
When, can also on the basis of the credit risk grade of user, with reference to pre-set business rule, class of subscriber, user tag,
The combined factors such as the current positioning address of user carry out credit risk scoring to it, and user's exhibition of preset requirement is only reached to scoring
Loan entrance is now preset, the excessive risk user that preset requirement is not reached to scoring does not show default loan entrance then, to prevent height
Risk subscribers handle loan transaction by default loan entrance.When showing default loan entrance, can scored according to device number
The terminal device for reaching the user of preset requirement is shown when this presets Application Program Interface in predeterminated position such as bottom, top etc.
APP interfaces key position shows default loan entrance, is provided a loan with facilitating guiding scoring to reach the user of preset requirement.
Compared with prior art, the present embodiment is carried out by the way that the seed user of different credit risk grades and user are drawn a portrait
Association, obtains user's representation data of different credit risk grades, and based on user's representation data of different credit risk grades
Training obtains the hierarchy model of credit risk grade, and the user in default application program is divided into difference using the hierarchy model
Credit risk grade；The different credit risk grades according to corresponding to user carry out risk score, only reach preset requirement to scoring
User show default loan entrance.Due to credit risk scoring can be carried out to the user preset in application program, only to scoring
Satisfactory user shows the loan entrance in default application program, improves to promote in the default application program of water conservancy diversion and borrows
The loan success rate of money business, the undesirable user client i.e. inferior that avoids scoring arbitrarily can also be handled by entrance of providing a loan
Loan, reduce the risk of finance company, the conversion ratio of lifting loan user.
In an optional embodiment, on the basis of above-described embodiment, the step S30 also includes：
Behavioral data of the user in default application program is obtained, according to the behavioral data by default application program
It is different behavior classifications that user, which carries out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is corresponding to credit risk grade
Weight coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
Behavioral data of the user in default application program can be also considered in the present embodiment, and according to the behavior number
It is different behavior classifications according to the user in default application program is carried out into cluster, for example, can be according to different user in default application
Consumer behavior data in program will be clustered often to consume, consuming once in a while, the classification such as not consuming substantially, or are accustomed to wholesale and are disappeared
Take, be accustomed to the classifications such as small amount consumption.According to practical application need to preset different credit risk grades corresponding to score base
Number M1 and its corresponding weight coefficient x1, and scoring radix M2 and its corresponding weight coefficient corresponding to different behavior classifications
Y1, COMPREHENSIVE CALCULATING obtain the risk score M of user.For example, predeterminable different credit risk grades are for example high, medium on the upper side, medium
Scoring radix is respectively 1,0.8,0.6,0.4 corresponding on the lower side, low, and weight coefficient x1 corresponding to credit risk grade is 0.4.In advance
If different behavior classifications are as often consumed, consumption, substantially not consumer other scoring radix are respectively 0.5,0.7,0.9 once in a while,
Weight coefficient corresponding to behavior classification is 0.6.If a user determines that it is high credit risk grade, and root in default application program
Determined that it is according to the behavioral data user of the user in default application program and consume classification once in a while, then the use can be calculated
The risk score M=0.4*1+0.6*0.7=0.82 at family.The risk that each user in default application program is calculated successively is commented
Point, the risk for the higher explanation loan that scores is higher, therefore, use of the preset requirement as being less than a certain predetermined threshold value is only reached to scoring
Family shows default loan entrance.
In the present embodiment, when calculating the risk score of user, the credit risk grade and use of user can be considered
Behavioral data of the family in default application program, the credit risk of user more accurately can be comprehensively evaluated, more accurately to enter
Row credit risk controls.
In an optional embodiment, the step S3 also includes：
Location data of the user in default application program is obtained, and according to the location data and the loan wind of the user
User's representation data corresponding to dangerous grade is to preset label corresponding to the user is set, then calculates user in default application program
Risk score M formula is：
M=x3*M3+y4*M4
Wherein, M3 is scoring radix corresponding to default different credit risk grades, and x3 is corresponding to credit risk grade
Weight coefficient；M4 is scoring radix corresponding to default different labels, and y4 is weight coefficient corresponding to label.
Location data of the user in default application program can be also considered in the present embodiment, and according to the positioning number
According to and the credit risk grade of the user corresponding to user's representation data be the user set corresponding to preset label.For example, can
It is luxury goods market according to the positioning address often occurred in the location data of user, and user's representation data corresponding to the user
In income attribute tags be booming income crowd, then can be the user set corresponding to preset label be " high consumption crowd "；If
Positioning address is common market, and the income attribute tags in user's representation data corresponding to the user are medium income group,
Can be then that label is preset corresponding to the user is set as " medium consumer groups ", etc..Set in advance according to the needs of practical application
Scoring radix M3 and its corresponding weight coefficient x3 corresponding to fixed different credit risk grades, and scored corresponding to different labels
Radix M4 and its corresponding weight coefficient y4, COMPREHENSIVE CALCULATING obtain the risk score M of user.For example, predeterminable different loan wind
Dangerous grade is for example high, it is medium it is on the upper side, it is medium it is on the lower side, low corresponding to scoring radix be respectively 1,0.8,0.6,0.4, credit risk grade
Corresponding weight coefficient x1 is 0.3.Preset the scoring of different labels such as high consumption crowd, medium consumer groups, low consumption crowd
Radix is respectively 0.3,0.5,0.9, and weight coefficient corresponding to behavior classification is 0.7.If a user determines in default application program
It is high credit risk grade, and determines that the label of the user is according to location data of the user in default application program
" medium consumer groups ", then the risk score M=0.3*1+0.5*0.7=0.65 of the user can be calculated.Calculate successively
The risk score of each user on to default application program, the risk for the higher explanation loan that scores is higher, and therefore, only scoring is reached
To preset requirement, such as user less than a certain predetermined threshold value shows default loan entrance.
In the present embodiment, when calculating the risk score of user, the credit risk grade and use of user can be considered
Location data of the family in default application program, the credit risk of user more accurately can be comprehensively evaluated, more accurately to enter
Row credit risk controls.
In addition, the present invention also provides a kind of computer-readable recording medium, the computer-readable recording medium storage has
Credit risk control system, the credit risk control system can be by least one computing devices, so that described at least one
The step of loan risk control method in computing device such as above-mentioned embodiment, the step S10 of the loan risk control method,
The specific implementation process such as S20, S30 as described above, will not be repeated here.
It should be noted that herein, term " comprising ", "comprising" or its any other variant are intended to non-row
His property includes, so that process, method, article or device including a series of elements not only include those key elements, and
And also include the other element being not expressly set out, or also include for this process, method, article or device institute inherently
Key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including this
Other identical element also be present in the process of key element, method, article or device.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can add the mode of required general hardware platform to realize by software, naturally it is also possible to realized by hardware, but a lot
In the case of the former be more preferably embodiment.Based on such understanding, technical scheme is substantially in other words to existing
The part that technology contributes can be embodied in the form of software product, and the computer software product is stored in a storage
In medium (such as ROM/RAM, magnetic disc, CD), including some instructions to cause a station terminal equipment (can be mobile phone, calculate
Machine, server, air conditioner, or network equipment etc.) perform method described in each embodiment of the present invention.
Above by reference to the preferred embodiments of the present invention have been illustrated, not thereby limit to the interest field of the present invention.On
State that sequence number of the embodiment of the present invention is for illustration only, do not represent the quality of embodiment.Patrolled in addition, though showing in flow charts
Order is collected, but in some cases, can be with the step shown or described by being performed different from order herein.
Those skilled in the art do not depart from the scope of the present invention and essence, can have a variety of flexible programs to realize the present invention,
It can be used for another embodiment for example as the feature of one embodiment and obtain another embodiment.All technologies with the present invention
The all any modification, equivalent and improvement made within design, all should be within the interest field of the present invention.
Claims (10)
1. a kind of electronic installation, it is characterised in that the electronic installation includes memory, processor, is stored on the memory
There is the credit risk control system that can be run on the processor, the credit risk control system is by the computing device
Shi Shixian following steps：
A, the seed user of default different credit risk grades is associated with the user's portrait pre-established, obtains difference
User's representation data of credit risk grade；
B, user's representation data based on different credit risk grades trains to obtain the hierarchy model of credit risk grade, and utilizes
The hierarchy model is classified to presetting the user in application program, is divided into different credit risk grades；
C, according to different credit risk grades corresponding to user in default application program, and by default code of points to default application
User in program carries out risk score, and the user that preset requirement is only reached to scoring shows default loan entrance.
2. electronic installation as claimed in claim 1, it is characterised in that credit risk grade includes four grades a, b, c, d, institute
The training process for stating hierarchy model is as follows：
User's representation data of different credit risk grades is trained using multiclass partitioning, when training set is extracted
Extract user's representation data vector corresponding to credit risk grade a respectively as positive to collect, corresponding to credit risk grade b, c, d
User's representation data vector as negative collection, to obtain the first training set；Extract user's portrait corresponding to credit risk grade b
Data vector extracts user's representation data vector corresponding to credit risk grade a, c, d as negative collection, to obtain as positive collection
Second training set；User's representation data vector corresponding to credit risk grade c is extracted as positive to collect, credit risk grade a, b,
User's representation data vector corresponding to d is as negative collection, to obtain the 3rd training set；Extract corresponding to credit risk grade d
As just collection, user's representation data vector corresponding to credit risk grade a, b, c collects user's representation data vector as negative, with
Obtain the 4th training set；This four training sets are trained respectively, four support vector machines are obtained, as credit risk
The hierarchy model of grade.
3. electronic installation as claimed in claim 2, it is characterised in that described to utilize the hierarchy model to presetting application program
On user carry out classification and include：
When being classified using the hierarchy model to the credit risk grade for presetting user in application program, by default application
Four support vector machines that the user data vector of user is utilized respectively in the hierarchy model in program are tested, and are obtained
Classification function value of the user data vector of user on each support vector machines on to default application program, will have most
Credit risk grade of the credit risk grade corresponding to the support vector machines of macrotaxonomy functional value as the user.
4. electronic installation as claimed in claim 1, it is characterised in that in the default application program of the basis corresponding to user not
The user's progress risk score preset in application program is included with credit risk grade, and by default code of points：
Behavioral data of the user in default application program is obtained, according to the behavioral data by the user in default application program
It is different behavior classifications to carry out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is weight corresponding to credit risk grade
Coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
5. a kind of loan risk control method, it is characterised in that the loan risk control method includes：
Step 1: the seed user of default different credit risk grades is associated with the user's portrait pre-established, obtain
To user's representation data of different credit risk grades；
Step 2: user's representation data based on different credit risk grades trains to obtain the hierarchy model of credit risk grade,
And be classified using the hierarchy model to presetting the user in application program, it is divided into different credit risk grades；
Step 3: the different credit risk grades according to corresponding to user in default application program, and by default code of points to pre-
If the user in application program carries out risk score, the user that preset requirement is only reached to scoring shows default loan entrance.
6. loan risk control method as claimed in claim 5, it is characterised in that credit risk grade include four grade a,
B, c, d, the training process of the hierarchy model are as follows：
User's representation data of different credit risk grades is trained using multiclass partitioning, when training set is extracted
Extract user's representation data vector corresponding to credit risk grade a respectively as positive to collect, corresponding to credit risk grade b, c, d
User's representation data vector as negative collection, to obtain the first training set；Extract user's portrait corresponding to credit risk grade b
Data vector extracts user's representation data vector corresponding to credit risk grade a, c, d as negative collection, to obtain as positive collection
Second training set；User's representation data vector corresponding to credit risk grade c is extracted as positive to collect, credit risk grade a, b,
User's representation data vector corresponding to d is as negative collection, to obtain the 3rd training set；Extract corresponding to credit risk grade d
As just collection, user's representation data vector corresponding to credit risk grade a, b, c collects user's representation data vector as negative, with
Obtain the 4th training set；This four training sets are trained respectively, four support vector machines are obtained, as credit risk
The hierarchy model of grade.
7. loan risk control method as claimed in claim 6, it is characterised in that described to utilize the hierarchy model to default
User in application program, which carries out classification, to be included：
When being classified using the hierarchy model to the credit risk grade for presetting user in application program, by default application
Four support vector machines that the user data vector of user is utilized respectively in the hierarchy model in program are tested, and are obtained
Classification function value of the user data vector of user on each support vector machines on to default application program, will have most
Credit risk grade of the credit risk grade corresponding to the support vector machines of macrotaxonomy functional value as the user.
8. loan risk control method as claimed in claim 5, it is characterised in that the basis presets user in application program
Corresponding different credit risk grades, and risk score bag is carried out to the user preset in application program by default code of points
Include：
Behavioral data of the user in default application program is obtained, according to the behavioral data by the user in default application program
It is different behavior classifications to carry out cluster, then the formula for calculating the risk score M of user in default application program is：
M=x1*M1+y1*M2
Wherein, M1 is scoring radix corresponding to default different credit risk grades, and x1 is weight corresponding to credit risk grade
Coefficient；M2 is scoring radix corresponding to default different behavior classifications, and y1 is weight coefficient corresponding to behavior classification.
9. loan risk control method as claimed in claim 5, it is characterised in that the basis presets user in application program
Corresponding different credit risk grades, and risk score bag is carried out to the user preset in application program by default code of points
Include：
Location data of the user in default application program is obtained, and according to credit risk of the location data and the user etc.
User's representation data corresponding to level is to preset label corresponding to the user is set, then calculates the risk of user in default application program
Scoring M formula be：
M=x3*M3+y4*M4
Wherein, M3 is scoring radix corresponding to default different credit risk grades, and x3 is weight corresponding to credit risk grade
Coefficient；M4 is scoring radix corresponding to default different labels, and y4 is weight coefficient corresponding to label.
10. a kind of computer-readable recording medium, it is characterised in that loan wind is stored with the computer-readable recording medium
Dangerous control system, realized when the credit risk control system is executed by processor as any one of claim 6 to 9
The step of loan risk control method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710916484.8A CN107767259A (en) | 2017-09-30 | 2017-09-30 | Loan risk control method, electronic installation and readable storage medium storing program for executing |
| PCT/CN2018/076139 WO2019061989A1 (en) | 2017-09-30 | 2018-02-10 | Loan risk control method, electronic device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710916484.8A CN107767259A (en) | 2017-09-30 | 2017-09-30 | Loan risk control method, electronic installation and readable storage medium storing program for executing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107767259A true CN107767259A (en) | 2018-03-06 |
Family
ID=61266987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710916484.8A Pending CN107767259A (en) | 2017-09-30 | 2017-09-30 | Loan risk control method, electronic installation and readable storage medium storing program for executing |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107767259A (en) |
| WO (1) | WO2019061989A1 (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108763051A (en) * | 2018-04-06 | 2018-11-06 | 平安证券股份有限公司 | Electronic device, transaction software operation risk method for early warning and storage medium |
| CN109345374A (en) * | 2018-09-17 | 2019-02-15 | 平安科技（深圳）有限公司 | Risk control method, device, computer equipment and storage medium |
| CN109584048A (en) * | 2018-11-30 | 2019-04-05 | 上海点融信息科技有限责任公司 | The method and apparatus that risk rating is carried out to applicant based on artificial intelligence |
| CN109766350A (en) * | 2018-12-17 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Partner's introduction method, device, computer equipment and storage medium |
| CN109859051A (en) * | 2019-01-15 | 2019-06-07 | 国网电子商务有限公司 | A kind of financing method and device |
| CN110135701A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Control automatic generation method, device, electronic equipment and the readable medium of rule |
| CN110163481A (en) * | 2019-04-19 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Electronic device, user's air control auditing system test method and storage medium |
| CN110197315A (en) * | 2018-04-08 | 2019-09-03 | 腾讯科技（深圳）有限公司 | Methods of risk assessment, device and its storage medium |
| CN110310012A (en) * | 2019-06-04 | 2019-10-08 | 平安科技（深圳）有限公司 | Data analysing method, device, equipment and computer readable storage medium |
| CN112948695A (en) * | 2021-03-31 | 2021-06-11 | 中国工商银行股份有限公司 | User portrait based general financial fast loan product recommendation method and device |
| CN113240509A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| WO2022105177A1 (en) * | 2020-11-17 | 2022-05-27 | 平安科技（深圳）有限公司 | User profile method and apparatus for credit card client, device, and medium |
| CN116245636A (en) * | 2023-03-03 | 2023-06-09 | 中科柏诚科技（北京）股份有限公司 | Loan risk decision method, device, equipment and medium based on user portrait |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110109905A (en) * | 2019-04-26 | 2019-08-09 | 深圳前海微众银行股份有限公司 | Risk list data generation method, device, equipment and computer storage medium |
| CN111062444B (en) * | 2019-12-21 | 2023-12-08 | 湖南大学 | Credit risk prediction method, credit risk prediction system, credit risk prediction terminal and storage medium |
| CN111260189B (en) * | 2020-01-08 | 2023-06-06 | 平安银行股份有限公司 | Risk control method, risk control device, computer system and readable storage medium |
| CN111383101B (en) * | 2020-03-25 | 2024-03-15 | 深圳前海微众银行股份有限公司 | Post-credit risk monitoring method, post-credit risk monitoring device, post-credit risk monitoring equipment and computer readable storage medium |
| CN111639706A (en) * | 2020-05-29 | 2020-09-08 | 深圳壹账通智能科技有限公司 | Personal risk portrait generation method based on image set and related equipment |
| CN112330321A (en) * | 2020-11-20 | 2021-02-05 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and computer readable storage medium |
| CN113436006A (en) * | 2021-07-07 | 2021-09-24 | 中国银行股份有限公司 | Loan risk prediction method and device based on block chain |
| CN113450209A (en) * | 2021-07-12 | 2021-09-28 | 中国银行股份有限公司 | Agricultural loan management and control system and method |
| CN113689289B (en) * | 2021-08-26 | 2024-04-30 | 天元大数据信用管理有限公司 | Bank risk control-based method and equipment |
| CN116074118B (en) * | 2023-03-07 | 2023-06-13 | 北京安胜华信科技有限公司 | API access control method, system, intelligent terminal and storage medium |
| CN116934464A (en) * | 2023-07-26 | 2023-10-24 | 广东企企通科技有限公司 | Post-loan risk monitoring method, device, equipment and medium based on small micro-enterprises |
| CN117217799A (en) * | 2023-09-20 | 2023-12-12 | 苏银凯基消费金融有限公司 | Financial user classification method based on user portrait and user behavior feature model |
| CN117350838A (en) * | 2023-10-27 | 2024-01-05 | 深圳市微云信众技术有限公司 | Bank shopping consumption coupon operation risk monitoring method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463673A (en) * | 2014-12-22 | 2015-03-25 | 中国科学技术大学苏州研究院 | P2P network credit risk assessment model based on support vector machine |
| KR101597939B1 (en) * | 2015-07-02 | 2016-02-25 | 한국기업데이터 주식회사 | Apparatus and method for predicting industrial credit risk using macro-economic indicator |
| CN106157122A (en) * | 2016-07-24 | 2016-11-23 | 国家电网公司 | Power consumer credit assessment method under a kind of internet environment |
| CN106557882A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | Power consumer screening technique and system based on various dimensions Risk Evaluation Factors |
| CN106897918A (en) * | 2017-02-24 | 2017-06-27 | 上海易贷网金融信息服务有限公司 | A kind of hybrid machine learning credit scoring model construction method |
| CN107038256A (en) * | 2017-05-05 | 2017-08-11 | 平安科技（深圳）有限公司 | Business customizing device, method and computer-readable recording medium based on data source |
| CN107203939A (en) * | 2017-05-26 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Determine method and device, the computer equipment of consumer's risk grade |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9064274B2 (en) * | 2011-08-04 | 2015-06-23 | Edward Y. Margines | Systems and methods of processing personality information |
| CN107203518A (en) * | 2016-03-16 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Method, system and device, the electronic equipment of on-line system personalized recommendation |
| CN106021376B (en) * | 2016-05-11 | 2019-05-10 | 上海点融信息科技有限责任公司 | Method and apparatus for handling user information |
| CN106503873A (en) * | 2016-11-30 | 2017-03-15 | 腾云天宇科技（北京）有限公司 | A kind of prediction user follows treaty method, device and the computing device of probability |
-
2017
- 2017-09-30 CN CN201710916484.8A patent/CN107767259A/en active Pending
-
2018
- 2018-02-10 WO PCT/CN2018/076139 patent/WO2019061989A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463673A (en) * | 2014-12-22 | 2015-03-25 | 中国科学技术大学苏州研究院 | P2P network credit risk assessment model based on support vector machine |
| KR101597939B1 (en) * | 2015-07-02 | 2016-02-25 | 한국기업데이터 주식회사 | Apparatus and method for predicting industrial credit risk using macro-economic indicator |
| CN106157122A (en) * | 2016-07-24 | 2016-11-23 | 国家电网公司 | Power consumer credit assessment method under a kind of internet environment |
| CN106557882A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | Power consumer screening technique and system based on various dimensions Risk Evaluation Factors |
| CN106897918A (en) * | 2017-02-24 | 2017-06-27 | 上海易贷网金融信息服务有限公司 | A kind of hybrid machine learning credit scoring model construction method |
| CN107038256A (en) * | 2017-05-05 | 2017-08-11 | 平安科技（深圳）有限公司 | Business customizing device, method and computer-readable recording medium based on data source |
| CN107203939A (en) * | 2017-05-26 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Determine method and device, the computer equipment of consumer's risk grade |
Non-Patent Citations (3)
| Title |
|---|
| 刘颖: "信用风险评级模型的应用研究", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
| 广东省江门市科学技术协会: "《江门市优秀科技论文集 2003-2004》", 31 December 2015, 广东省江门市科学技术协会 * |
| 熊建宇: "《互联网金融营销》", 30 April 2017, 中国金融出版社 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108763051A (en) * | 2018-04-06 | 2018-11-06 | 平安证券股份有限公司 | Electronic device, transaction software operation risk method for early warning and storage medium |
| CN108763051B (en) * | 2018-04-06 | 2023-07-25 | 平安证券股份有限公司 | Electronic device, transaction software running risk early warning method and storage medium |
| CN110197315A (en) * | 2018-04-08 | 2019-09-03 | 腾讯科技（深圳）有限公司 | Methods of risk assessment, device and its storage medium |
| CN110197315B (en) * | 2018-04-08 | 2024-03-22 | 腾讯科技（深圳）有限公司 | Risk assessment method, apparatus and storage medium thereof |
| CN109345374B (en) * | 2018-09-17 | 2023-04-18 | 平安科技（深圳）有限公司 | Risk control method and device, computer equipment and storage medium |
| CN109345374A (en) * | 2018-09-17 | 2019-02-15 | 平安科技（深圳）有限公司 | Risk control method, device, computer equipment and storage medium |
| CN109584048A (en) * | 2018-11-30 | 2019-04-05 | 上海点融信息科技有限责任公司 | The method and apparatus that risk rating is carried out to applicant based on artificial intelligence |
| CN109766350A (en) * | 2018-12-17 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Partner's introduction method, device, computer equipment and storage medium |
| CN109859051A (en) * | 2019-01-15 | 2019-06-07 | 国网电子商务有限公司 | A kind of financing method and device |
| CN110163481A (en) * | 2019-04-19 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Electronic device, user's air control auditing system test method and storage medium |
| CN110135701A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Control automatic generation method, device, electronic equipment and the readable medium of rule |
| CN110310012A (en) * | 2019-06-04 | 2019-10-08 | 平安科技（深圳）有限公司 | Data analysing method, device, equipment and computer readable storage medium |
| CN110310012B (en) * | 2019-06-04 | 2023-07-28 | 平安科技（深圳）有限公司 | Data analysis method, device, equipment and computer readable storage medium |
| WO2022105177A1 (en) * | 2020-11-17 | 2022-05-27 | 平安科技（深圳）有限公司 | User profile method and apparatus for credit card client, device, and medium |
| CN112948695A (en) * | 2021-03-31 | 2021-06-11 | 中国工商银行股份有限公司 | User portrait based general financial fast loan product recommendation method and device |
| CN113240509B (en) * | 2021-05-18 | 2022-04-22 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| CN113240509A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| CN116245636A (en) * | 2023-03-03 | 2023-06-09 | 中科柏诚科技（北京）股份有限公司 | Loan risk decision method, device, equipment and medium based on user portrait |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019061989A1 (en) | 2019-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107767259A (en) | Loan risk control method, electronic installation and readable storage medium storing program for executing | |
| CN107895277A (en) | Method, electronic installation and the medium of push loan advertisement in the application | |
| CN112148987B (en) | Message pushing method based on target object activity and related equipment | |
| CN105468742B (en) | The recognition methods of malice order and device | |
| CN109871446A (en) | Rejection method for identifying, electronic device and storage medium in intention assessment | |
| CN107730389A (en) | Electronic installation, insurance products recommend method and computer-readable recording medium | |
| CN110245213A (en) | Questionnaire generation method, device, equipment and storage medium | |
| CN110516910A (en) | Declaration form core based on big data protects model training method and core protects methods of risk assessment | |
| CN107465766A (en) | Information-pushing method, electronic equipment and computer-readable storage medium | |
| CN107808307A (en) | Business personnel's picture forming method, electronic installation and computer-readable recording medium | |
| CN107844634A (en) | Polynary universal model platform modeling method, electronic equipment and computer-readable recording medium | |
| CN107730310A (en) | Electronic installation, the method and storage medium for building Retail networks Rating Model | |
| CN108170759A (en) | Method, apparatus, computer equipment and the storage medium of tip-offs about environmental issues processing | |
| CN107682575A (en) | Business personnel's incoming call inlet wire distribution method, electronic installation, computer-readable recording medium | |
| CN107590291A (en) | A kind of searching method of picture, terminal device and storage medium | |
| CN107679084A (en) | Cluster labels generation method, electronic equipment and computer-readable recording medium | |
| CN105512773A (en) | Passenger travel destination prediction method and device | |
| CN108038655A (en) | Recommendation method, application server and the computer-readable recording medium of department's demand | |
| CN110309234A (en) | A kind of client of knowledge based map holds position method for early warning, device and storage medium | |
| CN106790570A (en) | A kind of consumer behaviour analysis and management system and its analysis method | |
| CN107688651A (en) | The emotion of news direction determination process, electronic equipment and computer-readable recording medium | |
| CN113051911A (en) | Method, apparatus, device, medium, and program product for extracting sensitive word | |
| CN103473275A (en) | Automatic image labeling method and automatic image labeling system by means of multi-feature fusion | |
| CN113688232A (en) | Method and device for classifying bidding texts, storage medium and terminal | |
| CN112507170A (en) | Data asset directory construction method based on intelligent decision and related equipment thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |
|
| RJ01 | Rejection of invention patent application after publication |