CN109377349A - A kind of accrediting amount evaluation method and device based on driving behavior - Google Patents
A kind of accrediting amount evaluation method and device based on driving behavior Download PDFInfo
- Publication number
- CN109377349A CN109377349A CN201811165478.4A CN201811165478A CN109377349A CN 109377349 A CN109377349 A CN 109377349A CN 201811165478 A CN201811165478 A CN 201811165478A CN 109377349 A CN109377349 A CN 109377349A
- Authority
- CN
- China
- Prior art keywords
- credit
- score
- user
- driving behavior
- information
- 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 OR CALCULATING; 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
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention discloses a kind of accrediting amount evaluation method and device based on driving behavior.Accrediting amount evaluation method of the one of which based on driving behavior, comprising: the first credit score is determined according to the first information of user, the accrediting amount is borrowed in the personal consumption that the first credit score is used to evaluate the user；Determine that the second credit score, the second credit score are used to evaluate the driving credit rating of the user according to the second information of the user；According to the first credit score and the second credit score, determine that the comprehensive credit line score of the user, the comprehensive credit line score are used to determine that the vehicle of the user to borrow the accrediting amount.Accrediting amount methods of marking based on driving behavior evaluates two kinds of scorings in conjunction with driving behavior evaluation and personal credit, for having vehicle to borrow client's more accurate positioning of demand while ensure that the basic interests of Che Dai company.
Description
Technical field
This application involves credit appraisal field more particularly to a kind of accrediting amount evaluation methods and dress based on driving behavior
It sets.
Background technique
Since entering 21 century, big data and internet finance have obtained huge development, peer-to-peer network (Peer-to-
Peer networking, P2P) one of important component as internet finance, it is advantageous using Internet technology
Advantage possesses characteristic more more convenient than traditional approach, quick and transparent, borrows field in small amount vehicle and rapidly develops, is to biography
Unite financial industry it is strong supplement and it is perfect.Che Dailei P2P company core is risk control, and personal credit scoring is then risk
The core of control.Personal credit scorecard is commonly constructed by promise breaking model at present, is based on obtaining from mechanisms such as banks
Information, the personal credit of user is evaluated, is evaluated according to the personal credit of user and determines loan limit.
But how constructing accuracy rate and can refer to the higher personal credit Rating Model of value is Che Dailei P2P company energy
The key point that no long-range business goes down, at present only only based on the information obtained from mechanisms such as banks, to the individual of user
Credit appraisal is apparently not comprehensive, location client, this evaluation method cannot not have for specified loaning bill reason accurately
Specific aim, it is disgraceful to will lead to last scoring, causes unnecessary economic loss to borrower's company.For example if one
A user's fund credit is good, but the driving behavior of the user is poor, and it is easy to appear traffic accidents for meeting, not only to inherently safe
It causes great threat while can also cause unnecessary economic loss to Che Dai company.
Summary of the invention
In view of the above problems, it proposes on the application overcomes the above problem or at least be partially solved in order to provide one kind
State a kind of accrediting amount evaluation method and device based on driving behavior of problem.
In a first aspect, the embodiment of the invention provides a kind of accrediting amount evaluation method based on driving behavior, it may include:
Determine that the first credit score, the first information include M credit data according to the first information of user, the first credit score is for commenting
The accrediting amount is borrowed in the personal consumption of valence user, and wherein M is the integer greater than 1；The second credit is determined according to the second information of user
Score, the second information include N class dangerous driving behavior data, and the second credit score is used to evaluate the driving credit of the user
Degree, N are the integer greater than 1；According to the first credit score and the second credit score, the comprehensive credit line score of user is determined,
Comprehensive credit line score is used to determine that the vehicle of the user to borrow the accrediting amount.
A kind of accrediting amount evaluation method based on driving behavior provided by first aspect, awarding based on driving behavior
Believe that amount points-scoring system substitutes personal credit scorecard model, is applied to vehicle and borrows in credit investigation system.By every often to debtor
The dangerous driving behavior index seen judge giving a mark and is dissolved into the scoring of its people's accrediting amount, in conjunction with driving behavior and individual
Two kinds of scorings of credit behavior, energy is more intuitive and accurately provides its comprehensive credit line amount, for there is vehicle to borrow the client of demand
More accurate positioning ensure that the basic interests of Che Dai company simultaneously.
In a kind of mode in the cards, the first credit score is determined according to the first information of user, may include: base
In feature branch mailbox method, by M credit data conversion at corresponding M weight evidence weight values woe；According to M weight evidence weight values woe and
Default individual's credit model determines the first credit score.
In a kind of mode in the cards, first is determined according to M weight evidence weight values woe and default personal credit model
Credit score may include: that M weight evidence weight values woe is updated in default personal credit model formation to determine the first credit
Score presets individual's credit model formation are as follows: score_1=∑ (woei*βi+α)*γ+θ；Wherein, score_1 is the first credit
Score；γ is the scale factor determined by the first preset formula；θ passes through the offset that the first preset formula determines；woeiTable
Show the corresponding weight evidence weight values woe of i-th of credit data in M credit data, wherein i=1,2 ..., M；βiBelieve for i-th
Borrow the regression coefficient that data are determined by preset Parameter Estimation Method；The α is time determined by preset Parameter Estimation Method
Return intercept.
In a kind of mode in the cards, the second credit score is determined according to the second information of user, may include: root
The credit score of the jth class dangerous driving behavior in the N class dangerous driving behavior data of user is determined according to the second information of user
score_2j, score_2jFor evaluating the driving credit rating of the jth class dangerous driving behavior of user；According to the second preset formulaDetermine the second credit score score_2, wherein j=1,2 ..., N.
In a kind of mode in the cards, according to the first credit score and the second credit score, comprehensive credit line point is determined
Number, may include: to determine comprehensive credit line score according to third preset formula score=δ * score_1+ (1- δ) * score_2；
Wherein, score_1 is the first credit score, and score_2 is the second credit score, and δ is preset threshold.
Second aspect, the embodiment of the invention provides a kind of accrediting amount evaluating apparatus based on driving behavior, it may include:
First credit unit, the second credit unit and comprehensive credit line unit.
First credit unit determines that the first credit score, the first information include M for the first information according to user
The accrediting amount is borrowed in a credit data, the personal consumption that the first credit score is used to evaluate the user, and wherein M is greater than 1
Integer；
Second credit unit determines the second credit score, second information for the second information according to the user
Including N class dangerous driving behavior data, the second credit score is used to evaluate the driving credit rating of the user, and N is greater than 1
Integer；
Comprehensive credit line unit, for determining the user according to the first credit score and the second credit score
Comprehensive credit line score, the comprehensive credit line score be used for determine the user vehicle borrow the accrediting amount.
In one possible implementation, the first credit unit may include: converting unit and determination unit.Wherein,
Converting unit, for being based on feature branch mailbox method, by M credit data conversion at corresponding M weight evidence weight values woe；It determines single
Member, for determining the first credit score according to M weight evidence weight values woe and default personal credit model.
In one possible implementation, determination unit is specifically used for, and M weight evidence weight values woe is updated to default
The first credit score is determined in personal credit model formation, presets individual's credit model formation are as follows: score_1=∑
(woei*βi+α)*γ+θ；Wherein, score_1 is the first credit score；γ is the ratio determined by the first preset formula
The factor；θ passes through the offset that the first preset formula determines；woeiIndicate i-th of credit data pair in the M credit data
The weight evidence weight values woe answered, wherein i=1,2 ..., M；βiIt is determined for i-th of credit data by preset Parameter Estimation Method
Regression coefficient；α is the recurrence intercept determined by preset Parameter Estimation Method.
In one possible implementation, the second credit unit can be specifically used for: true according to the second information of user
Determine the credit score score_2 of the jth class dangerous driving behavior in the N class dangerous driving behavior data of userj, score_2jWith
In the driving credit rating of the jth class dangerous driving behavior of evaluation user；According to the second preset formulaDetermine the second credit score score_2, wherein j=1,2 ..., N.
In one possible implementation, comprehensive credit line unit can be specifically used for: according to third preset formula score
=δ * score_1+ (1- δ) * score_2, determines comprehensive credit line score；Wherein, score_1 is the first credit score, score_2
For the second credit score, δ is preset threshold.
The third aspect, the embodiment of the invention provides a kind of computer readable storage medium, which is deposited
Program instruction is contained, the program instruction by processor when being run, the method which executes above-mentioned first aspect, herein not
It repeats again.
Fourth aspect, the embodiment of the invention provides a kind of accrediting amount evaluating apparatus based on driving behavior, including deposit
Component, processing component and communication component, storage assembly are stored up, processing component and communication component are connected with each other, wherein storage assembly is used
Code is handled in storing data, communication component is used to carry out information exchange with external equipment；Processing component is configured for calling
Program code executes method described in first aspect, and details are not described herein again.
It is apparently not complete to the personal credit evaluation of user at present only only based on the information obtained from mechanisms such as banks
Face, accurately location client, this evaluation method cannot there is no specific aim for specified loaning bill reason, will lead to last
It scores disgraceful, unnecessary economic loss is caused to borrower's company.Therefore, by doing woe volume after carrying out branch mailbox to feature
Code replacement, scorecard model is converted into using the interpretation of Logic Regression Models, forms personal accrediting amount points-scoring system,
The first information with access customer obtains the first credit score of user.The dangerous driving behavior data of default processing user, obtain
User's the second credit score, comprehensive first credit score and the second credit score obtain comprehensive credit line score, and this combination drives
Behavior and personal credit two kinds of scorings of behavior, energy is more intuitive and accurately provides its comprehensive credit line amount, for specified
Loaning bill reason has strong specific aim, preferably ensure that the accuracy finally to score, more smart for the client's positioning for having vehicle to borrow demand
Standard has evaded the unnecessary economic loss caused by borrower's company.
Detailed description of the invention
Technical solution in order to illustrate the embodiments of the present invention more clearly or in background technique below will be implemented the present invention
Attached drawing needed in example or background technique is illustrated.
Fig. 1 is a kind of accrediting amount evaluation system frame diagram based on driving behavior provided in an embodiment of the present invention；
Fig. 2 is a kind of signal of accrediting amount evaluation method process based on driving behavior provided in an embodiment of the present invention
Figure；
Fig. 3 is the signal of another accrediting amount evaluation method process based on driving behavior provided in an embodiment of the present invention
Figure；
Fig. 4 is a kind of structural representation of accrediting amount evaluating apparatus based on driving behavior provided in an embodiment of the present invention
Figure；
Fig. 5 is the structural representation of another accrediting amount evaluating apparatus based on driving behavior provided in an embodiment of the present invention
Figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and the embodiment of the present invention is described.
The description and claims of this application and term " first " in the attached drawing, " second " and " third " etc. are
For distinguishing different objects, it is not use to describe a particular order.In addition, term " includes " and " having " and their any changes
Shape, it is intended that cover and non-exclusive include.Such as contain the process, method of a series of steps or units, system, product or
Equipment is not limited to listed step or unit, but optionally further comprising the step of not listing or unit or optional
Ground further includes the other step or units intrinsic for these process, methods, product or equipment.
Referenced herein " embodiment " is it is meant that a particular feature, structure, or characteristic described can wrap in conjunction with the embodiments
It is contained at least one embodiment of the application.Each position in the description occur the phrase might not each mean it is identical
Embodiment, nor the independent or alternative embodiment with other embodiments mutual exclusion.Those skilled in the art explicitly and
Implicitly understand, embodiment described herein can be combined with other embodiments.
Firstly, the part term in the application is explained, in order to those skilled in the art understand that.
(1) logistic regression algorithm, logistic regression (logistic) are a kind of generalized linear regression (generalized
Linear model), therefore have many something in common with multiple linear regression analysis.Their model form is substantially the same,
All there is w ' x+b, wherein w and b is parameter to be asked, and difference is that their dependent variable is different, and multiple linear regression directly will
W ' x+b is as dependent variable, i.e. y=w ' x+b, and logistic is returned and w ' x+b is then corresponded to a hidden state p, p by function L
=L (w ' x+b) then determines the value of dependent variable according to the size of p and 1-p.
(2) regression coefficient, regression coefficient (regression coefficient) indicate x pairs of independent variable in regression equation
Dependent variable y influences the parameter of size, and the bigger expression x of regression coefficient influences y bigger.Regression coefficient can be by the way of estimation
It is calculated.
(3) evidence weight (Weight of Evidence, WOE) scorecard model, principle is to encode model variable woe
The generalized linear model for two classified variable of one kind for using Logic Regression Models to carry out after mode discretization.Seek the public affairs of woe value
Formula is exactly: woe=ln (hospitable family accounting/bad client accounting) * 100%=odds ratio.Can be divided into application scorecard (A card,
Application scorecard), B card (Behavior scorecard) and C card (Collection scorecard), point
It Yong Yu not post-loan management and collection management.
(4) feature branch mailbox method, feature branch mailbox method refer to through investigation " neighbours " (value of surrounding) come smooth storing data
Value indicates taking for each bin values with " width of case " with the data for having same number in the different case of " depth of case " expression
It is worth section.It is a kind of local smoothing method method since branch mailbox method considers adjacent value.The main purpose of branch mailbox is denoising,
By continuous data discretization, increase granularity.According to the difference of value can be divided into it is smooth by case average value, by case median smoothing with
And it is smooth by case boundary value.Feature branch mailbox method can be divided into unsupervised branch mailbox method and have supervision branch mailbox method.Unsupervised branch mailbox method can
To be divided into wide branch mailbox method, frequency divisions case method is waited, the branch mailbox method etc. based on k mean cluster.There is supervision branch mailbox method that can be divided into card side
Branch mailbox method etc..
Secondly, one of accrediting amount evaluation system framework based on driving behavior being based on to the embodiment of the present invention
It is described, refers to attached drawing 1, Fig. 1 is a kind of system architecture schematic diagram of accrediting amount based on driving behavior, such as Fig. 1 institute
Show, the accrediting amount evaluation method based on driving behavior that the application proposes can be applied to the system architecture.The system architecture
Contain the contents such as vehicle 1, vehicle 2, vehicle drive behavioral value equipment and background server.
Vehicle 1 indicates the first information that vehicle 1 is uploaded without vehicle drive behavioral value appliance icon beside vehicle 1
It can be to pass through vehicle including the dangerous driving behavior information that cannot be directly or indirectly detected by vehicle drive behavioral value equipment
Identification information (identification information can be used to unique identification target vehicle driving behavior information, for example: identification information
Can be license plate number) or user identification information (identification information can be used to unique identification user driving behavior information,
For example: identification information can be ID card No.) identify the dangerous driving behavior that user is corresponded in traffic department's record
Information.
Vehicle 2 and vehicle drive behavioral value equipment indicate that the first information that vehicle 2 is uploaded includes that can be driven by vehicle
Sail the dangerous driving behavior information that behavioral value equipment directly or indirectly detects.Due to making for vehicle drive behavioral value equipment
Mostly inside the vehicle with position, so vehicle drive behavioral value appliance icon is connected beside vehicle with vehicle；Vehicle is driven
Sail behavioral value equipment detection dangerous driving behavior information can be uploaded to by wired or wireless communications mode after
Platform server.
Background server, the icon of background server represents background server and can be to be made of several servers.Afterwards
For determining that the first credit score, the first information include M credit data according to the first information of user, first awards platform server
Letter score is used to evaluate the consumer credit credit of user, and wherein M is the integer greater than 1；It is true according to the second information of user
Fixed second credit score, the second information include N class dangerous driving behavior data, and the second credit score is used to evaluate the driving of user
Credit rating, N are the integer greater than 1；According to the first credit score and the second credit score, the comprehensive credit line score of user is determined,
Comprehensive credit line score is used to determine that the vehicle of user to borrow the accrediting amount.Background server can be also used for obtaining the first information and second
Information.
It is understood that the system architecture in Fig. 1 is the illustrative embodiment of one of embodiment of the present invention.
System architecture in the embodiment of the present invention may include but be not limited only to system above framework.
Attached drawing 2 is please referred to, Fig. 2 is a kind of accrediting amount evaluation method based on driving behavior provided in an embodiment of the present invention
The schematic diagram of process.It can be applied to the system in above-mentioned Fig. 1, this method may comprise steps of S201- step S203.
Step S201: the first credit score is determined according to the first information of user.
Specifically, the first information includes M credit data, and the personal consumption loan that the first credit score is used to evaluate user is awarded
Believe amount, wherein M is the integer greater than 1.The first credit score of user, the first credit point of user are determined according to credit data
Number illustrates the personal credit of user, and the accrediting amount is borrowed in the personal consumption for evaluating the user.The personal consumption of user is borrowed
The accrediting amount can be bank or other financial institutions and provide the commodity-type currency of quota to user.
Optionally, credit data may is that Unionpay's data, internet behavioral data, public sentiment data etc..For example, silver-colored
Connection data can be whether user has loan documentation, if having overdue record of debt-credit etc.；Internet behavioral data can be user
All behaviors occurred on website；Public sentiment data can be public sentiment and occur because becoming item, in development and change procedure, Min Zhongsuo
The social attitude held.
Optionally, consumer credit can be bank or other financial institutions take credit, mortgage, pledge guarantee or guarantor
Card mode, the credit provided with commodity-type money-form to individual consumer.
Optionally, according to the score value of the first credit score, personal consumption can be borrowed the accrediting amount and is chosen as four grades.
Step S202: the second credit score is determined according to the second information of user.
Specifically, the second information includes N class dangerous driving behavior data, and the second credit score is for evaluating the user's
Credit rating is driven, N is that the integer greater than 1 determines the second credit score, Neng Gouping according to the dangerous driving behavior data of user
Estimate the driving safety risk of user.
Optionally, N class dangerous driving behavior data may include: described in N class dangerous driving behavior categorical data and generation
The corresponding time data of N class dangerous driving behavior, wherein optionally, N class dangerous driving behavior categorical data may include: average
Oil consumption, the alarm of sudden turn of events speed, idling time duration, overspeed alarming, mileage maximum speed, hypervelocity duration etc. can direct or indirect bases
The dangerous driving behavior information of mobile unit detection；Optional N class dangerous driving behavior categorical data, also may include: traffic
Drunk driving information in system, illegal parking information, overcrowding loaded information etc. can not be directly or indirectly according to mobile unit detections
Dangerous driving behavior information.It can be understood that N class dangerous driving behavior categorical data is also possible to all danger that can be obtained
Dangerous driving behavior information, including can directly or indirectly according to mobile unit detection dangerous driving behavior information and can not directly or
The dangerous driving behavior information detected indirectly according to mobile unit.
Optionally, the second information can also include, the directly or indirectly obtained according to vehicle drive behavioral value equipment
Two information；Alternatively, the second information obtained according to the vehicles identifications of user or user identifier in related traffic department.
Step S203: according to the first credit score and the second credit score, the comprehensive credit line score of user is determined.
Specifically, comprehensive credit line score is used to determine that the vehicle of the user to borrow the accrediting amount.In conjunction with the first credit score (root
It is believed that borrowing the first determining credit score) and (the second credit point determined according to dangerous driving behavior data of the second credit score
Number) comprehensive credit line score is determined jointly.For example: when the first credit score is 800 points and the second credit score is 200 points
When, comprehensive credit line score can be 800*0.8+ (1-0.8) * 200=680, wherein 0.8 is preset threshold.According to comprehensive credit line
Score 680 can be to the 70% of user's quota loan.
Implement the embodiment of the present invention, personal credit evaluation system can be substituted based on the accrediting amount points-scoring system of driving behavior
System is applied to vehicle and borrows in credit investigation system.By scoring the every common dangerous driving behavior index of debtor, then with its
The scoring of the personal accrediting amount combines, this combination driving behavior and personal credit two kinds of scorings of behavior, can it is more intuitive and
Accurately provide its comprehensive credit line score, comprehensive credit line score is compared to only with that personal accrediting amount scoring is in the present embodiment
One credit score, vehicle credit side face of the meeting preferably suitable for personal consumption loan, positions not only for the client for having vehicle to borrow demand
The basic interests of Che Dai company are more precisely also ensured simultaneously.
Attached drawing 3 is please referred to, Fig. 3 is the accrediting amount evaluation method process provided in an embodiment of the present invention based on driving behavior
Schematic diagram, this method may comprise steps of S301- step S305.
Step S301: it is based on feature branch mailbox method, by M credit data conversion at corresponding M weight evidence weight values woe.
Specifically, feature branch mailbox method may include unsupervised branch mailbox method and have supervision branch mailbox method.Unsupervised branch mailbox method can be with
It is divided into wide branch mailbox method, waits frequency divisions case method, the branch mailbox method etc. based on k mean cluster.There is supervision branch mailbox method that can be divided into card side point
Case method etc..
Optionally, before above-mentioned steps S301, the first information of user can also be obtained, the first information includes M letter
Data are borrowed, wherein M is optional greater than 1 integer, and credit data may is that Unionpay's data, internet behavioral data, public sentiment number
According to etc..
Step 302: the first credit score is determined according to M weight evidence weight values woe and default personal credit model.
Specifically, the M weight evidence weight values woe is updated in default personal credit model formation and determines described first
Credit score, the default personal credit model formation are as follows: score_1=∑ (woei*βi+α)*γ+θ；Wherein, described
Score_1 is the first credit score；The γ is the scale factor determined by the first preset formula；The θ passes through institute
State the offset that the first preset formula determines；The woeiIndicate that i-th of credit data in the M credit data are corresponding
Weight evidence weight values woe, wherein i=1,2 ..., M；The βiIt is determined for i-th of credit data by preset Parameter Estimation Method
Regression coefficient；The α is the recurrence intercept determined by the preset Parameter Estimation Method.
Optionally, the first preset formula includes: score_0=log (X) * γ+θ and score_0+PDO=log (2X) * γ
+θ；Wherein, the score_0 is basic score value that is known or assuming；The X is the fine or not sample ratio that is known or assuming
Rate；The PDO is the double score value of the fine or not sample ratio that is known or assuming.Wherein fine or not sample ratio preferably client
(there is no or less that there is a situation where personal credits is overdue) (it is overdue that personal credit will often occur, or for a long time with bad client
It is overdue) ratio of number
Optionally, preset Parameter Estimation Method may include: the sum of likelihood function maximization method, error minimization or minimum
Two multiply the Parameter Estimation Method of the Logic Regression Models such as estimation.For example: in a linear regression model (LRM), unknown regression coefficient
The least-squares estimation of β is to meet minβ||Y-Xβ||2β.Know that β is equation (XTX) β=XTThe solution of Y.This equation is known as just
Advise equation.Due in linear regression model (LRM), X matrix sequency spectrum, therefore β can be released, and be denoted as β=(XTX)-1XTY。
Optionally, woe value represents fine or not sample ratio, alternatively, woe value indicates the risk assessment of bad client in branch mailbox, woe value
It is higher that represent bad customer risk in branch mailbox lower.Reach preset condition (such as: without overdue record in bank loan information) preferably
Client, not up to preset condition are bad client.
Optionally, α=α1+α2+…+αi+…+αM, wherein αiPass through preset Parameter Estimation Method for i-th of credit data
Determining recurrence intercept.
Optionally, credit is borrowed in the first credit fraction representation personal credit of user, the personal consumption for evaluating user
Amount.The personal consumption of user borrow the accrediting amount can be with consumer credit (consumer credit can be bank or its
His financial institution takes credit, mortgages, pledges guarantee or guarantee mode, is provided with commodity-type money-form to individual consumer
Credit) form to user provide quota commodity-type currency.
Optionally, according to the score value of the first credit score, personal consumption can be borrowed the accrediting amount and is chosen as four grades, lifted
For example: can be divided into four score sections according to the size of score, each score section respectively correspond it is excellent, it is good, it is poor, very poor four etc.
Grade.
Step S303: determine that the jth class in the N class dangerous driving behavior data of user is dangerous according to the second information of user
The credit score score_2 of driving behaviorj。
Specifically, score_2jFor evaluating the driving credit rating of the jth class dangerous driving behavior of the user.
Optionally, before above-mentioned steps S303, the second information of user can also be obtained, the second information includes N class danger
Dangerous driving behavior data, N are the integer greater than 1.
Optionally, the mode for obtaining the second information of user can also include, directly or indirectly according to vehicle drive behavior
Detection device obtains；Alternatively, being obtained according to the vehicles identifications of user or user identifier in related traffic department.
Optionally, determine that the jth class danger in the N class dangerous driving behavior data of user is driven according to the second information of user
Sail the credit score score_2 of behaviorj, may include: by counting the distribution of quartile point to N class dangerous driving behavior data
To calculate the credit score score_2 of jth class dangerous driving behaviorj.For example: can be for jth class dangerous driving behavior
Number take 9 sections from 0-7 and greater than 7, with the increase of dangerous driving behavior number, jth class dangerous driving behavior
Credit score score_2jScore value be gradually reduced, be greater than 7 when score_2jScore value be set as minimum -5, be equal to 0 when
score_2jScore value be set as highest 25, the credit score score_2 of jth class dangerous driving behavior at this timejScore is higher, jth
Class dangerous driving behavior number is fewer, and driving behavior is better.
Optionally, determine that the jth class danger in the N class dangerous driving behavior data of user is driven according to the second information of user
Sail the credit score score_2 of behaviorj, can also include: to N class dangerous driving behavior data jth class dangerous driving behavior
J class dangerous driving behavior authorizes basic score value each time, carries out to each data score value of jth class dangerous driving behavior tired
Add, obtains the credit score score_2 of jth class dangerous driving behaviorj.That is score_2jThe score value * jth class dangerous driving of=basis
Behavior number.The credit score score_2 of jth class dangerous driving behavior at this timejScore is lower, jth class dangerous driving behavior time
Number is fewer, and driving behavior is better.
Optionally, determine that the jth class danger in the N class dangerous driving behavior data of user is driven according to the second information of user
Sail the credit score score_2 of behaviorj, can also include: to N class dangerous driving behavior data jth class dangerous driving behavior
J class dangerous driving behavior authorizes basic score value each time, to each data score value and jth class of jth class dangerous driving behavior
Dangerous driving behavior shared multiplied by weight in N class dangerous driving behavior, then add up, obtain jth class the second credit score,
That is score_2j=basis score value * jth class dangerous driving behavior shared weight in N class dangerous driving behavior.Jth class is endangered at this time
The credit score score_2 of dangerous driving behaviorjScore is lower, and driving behavior is better.
Step 304: the second credit score score_2 is determined according to the second preset formula.
Specifically, the second preset formula can beWherein j=1,2 ..., N.The
The driving credit rating that two credit scores are used to evaluate the user determines that second awards according to the dangerous driving behavior data of user
Believe score, the driving safety risk of user can be assessed.Second credit score score_2 is higher, and driving behavior is better.
Optionally, if the credit score score_2 of jth class dangerous driving behaviorjScore is lower, jth class dangerous driving row
Fewer for number, driving behavior is better, then the second preset formula can be with are as follows: Wherein basic total score can be total for preset N class dangerous driving behavior data
Score value.
Step S305: according to third preset formula score=δ * score_1+ (1- δ) * score_2, the synthesis is determined
Credit score.
Specifically, score_1 is the first credit score, and score_2 is the second credit score, and δ is default threshold
Value, comprehensive credit line score score are used to determine that the vehicle of user to borrow the accrediting amount.It (is determined according to credit in conjunction with the first credit score
The first credit score) and the second credit score (the second credit score determined according to dangerous driving behavior data) is common determines
Comprehensive credit line score, comprehensive credit line score compared to only use the first credit score, preferably be suitable for personal consumption borrow vehicle loan.
Optionally, comprehensive credit line score is used to determine that the accrediting amount of the user to include: when comprehensive credit line point reaches mark
When quasi- value, comprehensive credit line score is higher, user obtain the accrediting amount is bigger or user obtain amount chance it is bigger；Or
Person does not give user's accrediting amount when comprehensive credit line score is less than standard value.
It optionally, can appropriate adjustment preset threshold δ size in certain scene.For example: when multiple first credits
When score section where score corresponds to the user of top grade while applying for that vehicle is borrowed, it can suitably reduce preset threshold δ, so that in synthesis
When evaluation determines comprehensive credit line score, driving behavior i.e. the second credit score of user is more considered.
Woe coding replacement is done after carrying out branch mailbox to feature through the embodiment of the present invention, utilizes solving for Logic Regression Models
The property released is converted into scorecard model, forms personal accrediting amount points-scoring system, the first information with access customer obtains user's
First credit score.The dangerous driving behavior data of default processing user, obtain user's the second credit score, the second credit score
It is lower, it was demonstrated that the driving behavior of user is poorer, and the driving behavior the poor more is easy to happen traffic accident, is both to oneself life
Irresponsibility be also a kind of threat to other people life.Comprehensive first credit score and the second credit score obtain comprehensive credit line
Score, this combination driving behavior and personal credit two kinds of scorings of behavior, can be more intuitive and accurately provide its synthesis and award
Believe amount, have strong specific aim for specified loaning bill reason, preferably ensure that the accuracy finally to score, is needed for there is vehicle to borrow
The client's more accurate positioning asked has evaded the unnecessary economic loss caused by borrower's company.
Attached drawing 4 is please referred to, Fig. 4 is a kind of accrediting amount evaluating apparatus based on driving behavior provided in an embodiment of the present invention
Structural schematic diagram.For example, evaluating apparatus can be it is a kind of by quick obtaining, processing, analysis and extract it is valuable,
Magnanimity and enriched data bring various convenient service equipments based on interaction data for third party's use.It can also join
Examine background server shown in FIG. 1.It may include the first credit unit 401 in Fig. 4 device 40, the second credit unit 402 is comprehensive
Credit unit 403 is closed, can also include converting unit 404 and determination unit 405.
First credit unit 401 determines the first credit score, the first information packet for the first information according to user
M credit data are included, the accrediting amount is borrowed in the personal consumption that the first credit score is used to evaluate the user, and wherein M is big
In 1 integer；
Optionally, the first credit unit 401 includes: converting unit 404 and determination unit 405, and converting unit 404 is used for
Based on feature branch mailbox method, by M credit data conversion at corresponding M weight evidence weight values woe；
Determination unit 405, for determining the first credit point according to M weight evidence weight values woe and default personal credit model
Number.
Optionally, determination unit 405, it is public specifically for M weight evidence weight values woe is updated to default personal credit model
The first credit score is determined in formula, presets individual's credit model formation are as follows: score_1=∑ (woei*βi+α)*γ+θ；Wherein,
Score_1 is the first credit score；γ is the scale factor determined by the first preset formula；θ is true by the first preset formula
Fixed offset；woeiIndicate the corresponding weight evidence weight values woe of i-th of credit data in M credit data, wherein i=1,
2 ..., M；βiThe regression coefficient determined for i-th of credit data by preset Parameter Estimation Method；α is to pass through preset parameter
The recurrence intercept that the estimation technique determines.
Second credit unit 402 determines the second credit score, second letter for the second information according to the user
Breath includes N class dangerous driving behavior data, and the second credit score is used to evaluate the driving credit rating of the user, and N is big
In 1 integer；
Optionally, the second credit unit 402 determines that the N class danger of user is driven specifically for the second information according to user
Sail the credit score score_2 of the jth class dangerous driving behavior in behavioral dataj, score_2jFor evaluating the jth class of user
The driving credit rating of dangerous driving behavior；According to the second preset formula Determine the second credit
Score score_2, wherein j=1,2 ..., N.
Comprehensive credit line unit 403, for determining the comprehensive credit line of user according to the first credit score and the second credit score
Score, comprehensive credit line score are used to determine that the vehicle of user to borrow the accrediting amount.
Optionally, comprehensive credit line unit 403, is specifically used for: according to third preset formula score=δ * score_1+ (1-
δ) * score_2 determines comprehensive credit line score；Wherein, score_1 is the first credit score, and score_2 is the second credit score,
δ is preset threshold.
It should be noted that the realization of each operation can also be to the phase that should refer to Fig. 2, embodiment of the method shown in Fig. 3
It should describe, details are not described herein again.
Attached drawing 5 is please referred to, Fig. 5 is another accrediting amount evaluation dress based on driving behavior provided in an embodiment of the present invention
The structural schematic diagram set, it may include following one or more components in Fig. 5 device 50 that it is convenient, which to be easy to understand and illustrate: storage
Component 501, processing component 502, communication component 503.
Storage assembly 501 may include one or more storage units, and each unit may include one or more storages
Device, storage assembly can be used for storing program and various data, and can in 50 operational process high speed of device, be automatically completed program
Or the access of data.Information, described two stable states point can be stored using the physical device of stable state there are two types of tools
It is not expressed as " 0 " and " 1 ".When device 50 shown in fig. 5, when being background server described in Fig. 1, storage assembly can be used to store
The first information, the second information and other related datas etc..
Processing component 502, processing component are referred to as processor, and processing unit handles veneer, processing module, processing
Device etc..Processing component can be central processing unit (central processing unit, CPU), network processing unit
The combination of (network processor, NP) or CPU and NP.It is background server described in Fig. 1 when device 50 shown in fig. 5
When, the processing component 503 is for calling the data of the storage assembly 501 to execute the correlation of above-mentioned Fig. 2 to Fig. 3 the method
Description, details are not described herein again.
Communication component 503 is referred to as transceiver or transceiver etc., wherein may include wireless, wired for carrying out
Or the unit of other communication modes.Optionally, it is single that the device in 503 parts for realizing receive capabilities can be considered as to reception
Member will be considered as transmission unit for realizing the device of sending function, i.e. 503 parts include receiving unit and matching unit.
It should be noted that the specific implementation of each operation can also be to should refer to Fig. 2, embodiment of the method shown in Fig. 3
Corresponding description, details are not described herein again.
In this application, the unit as illustrated by the separation member may or may not be physically separate
, component shown as a unit may or may not be physical unit, it can and it is in one place, or can also
To be distributed over a plurality of network elements.Some or all of unit therein can be selected to realize this hair according to the actual needs
The purpose of bright example scheme.
In addition, each functional unit in each embodiment of the application, which can integrate, is also possible to each group in a component
Part physically exists alone, and is also possible to two or more components and is integrated in a component.Above-mentioned integrated component both may be used
To use formal implementation of hardware, can also realize in the form of software functional units.
If the integrated component is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the application
Portion or part steps.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should all cover within the scope of protection of this application.Therefore, the protection scope of the application should be with right
It is required that protection scope subject to.
It should be understood that magnitude of the sequence numbers of the above procedures are not meant to execute suitable in the various embodiments of the application
Sequence it is successive, the execution of each process sequence should be determined by its function and internal logic, the implementation without coping with the embodiment of the present invention
Process constitutes any restriction.Although the application is described in conjunction with each embodiment herein, however, being protected required by embodiment
During the application of shield, those skilled in the art are appreciated that and realize other variations of the open embodiment.
Claims (10)
1. a kind of accrediting amount evaluation method based on driving behavior characterized by comprising
Determine that the first credit score, the first information include M credit data according to the first information of user, described first awards
The accrediting amount is borrowed in the personal consumption that letter score is used to evaluate the user, and wherein M is the integer greater than 1；
The second credit score is determined according to the second information of the user, and second information includes N class dangerous driving behavior number
According to the second credit score is used to evaluate the driving credit rating of the user, and N is the integer greater than 1；
According to the first credit score and the second credit score, the comprehensive credit line score of the user is determined, it is described comprehensive
It closes credit score and is used to determine that the vehicle of the user to borrow the accrediting amount.
2. method according to claim 1, which is characterized in that the first information according to user determines the first credit point
Number, comprising:
Based on feature branch mailbox method, by the M credit data conversion at corresponding M weight evidence weight values woe；
The first credit score is determined according to the M weight evidence weight values woe and default personal credit model.
3. method according to claim 2, which is characterized in that described according to the M weight evidence weight values woe and default individual
Credit model determines the first credit score, comprising:
The M weight evidence weight values woe is updated in default personal credit model formation and determines the first credit score, institute
State default personal credit model formation are as follows: score_1=∑ (woei*βi+α)*γ+θ；Wherein, the score_1 is described the
One credit score；The γ is the scale factor determined by the first preset formula；The θ is true by first preset formula
Fixed offset；The woeiIndicate the corresponding weight evidence weight values woe of i-th of credit data in the M credit data,
Middle i=1,2 ..., M；The βiThe regression coefficient determined for i-th of credit data by preset Parameter Estimation Method；The α is
The recurrence intercept determined by the preset Parameter Estimation Method.
4. method according to claim 3, which is characterized in that described to determine the second credit according to the second information of the user
Score, comprising:
The jth class dangerous driving in the N class dangerous driving behavior data of the user is determined according to the second information of the user
The credit score score_2 of behaviorj, the score_2jThe jth class dangerous driving behavior for evaluating the user is driven
Sail credit rating；
According to the second preset formulaDetermine the second credit score score_2, wherein j
=1,2 ..., N.
5. method according to claim 4, which is characterized in that described according to the first credit score and second credit
Score determines comprehensive credit line score, comprising:
According to third preset formula score=δ * score_1+ (1- δ) * score_2, the comprehensive credit line score is determined；Wherein,
Score_1 is the first credit score, and score_2 is the second credit score, and δ is preset threshold.
6. a kind of accrediting amount evaluating apparatus based on driving behavior characterized by comprising
First credit unit determines that the first credit score, the first information include M letter for the first information according to user
Data are borrowed, the accrediting amount is borrowed in the personal consumption that the first credit score is used to evaluate the user, and wherein M is whole greater than 1
Number；
Second credit unit determines the second credit score for the second information according to the user, and second information includes N
Class dangerous driving behavior data, the second credit score are used to evaluate the driving credit rating of the user, and N is whole greater than 1
Number；
Comprehensive credit line unit, for determining that the user's is comprehensive according to the first credit score and the second credit score
Credit score is closed, the comprehensive credit line score is used to determine that the vehicle of the user to borrow the accrediting amount.
7. method according to claim 6, which is characterized in that the first credit unit includes:
Converting unit, for being based on feature branch mailbox method, by the M credit data conversion at corresponding M weight evidence weight values
woe；
Determination unit, for determining first credit point according to the M weight evidence weight values woe and default personal credit model
Number.
8. device according to claim 7, which is characterized in that the determination unit is specifically used for:
The M weight evidence weight values woe is updated in default personal credit model formation and determines the first credit score, institute
State default personal credit model formation are as follows: score_1=∑ (woei*βi+α)*γ+θ；Wherein, the score_1 is described the
One credit score；The γ is the scale factor determined by the first preset formula；The θ is true by first preset formula
Fixed offset；The woeiIndicate the corresponding weight evidence weight values woe of i-th of credit data in the M credit data,
Middle i=1,2 ..., M；The βiThe regression coefficient determined for i-th of credit data by preset Parameter Estimation Method；The α is
The recurrence intercept determined by the preset Parameter Estimation Method.
9. device according to claim 8, which is characterized in that the second credit unit is specifically used for:
The jth class dangerous driving in the N class dangerous driving behavior data of the user is determined according to the second information of the user
The credit score score_2 of behaviorj, the score_2jThe jth class dangerous driving behavior for evaluating the user is driven
Sail credit rating；
According to the second preset formulaDetermine the second credit score score_2, wherein j
=1,2 ..., N.
10. device according to claim 9, which is characterized in that the comprehensive credit line unit is specifically used for:
According to third preset formula score=δ * score_1+ (1- δ) * score_2, the comprehensive credit line score is determined；Wherein,
Score_1 is the first credit score, and score_2 is the second credit score, and δ is preset threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811165478.4A CN109377349A (en) | 2018-09-30 | 2018-09-30 | A kind of accrediting amount evaluation method and device based on driving behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811165478.4A CN109377349A (en) | 2018-09-30 | 2018-09-30 | A kind of accrediting amount evaluation method and device based on driving behavior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109377349A true CN109377349A (en) | 2019-02-22 |
Family
ID=65403649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811165478.4A Pending CN109377349A (en) | 2018-09-30 | 2018-09-30 | A kind of accrediting amount evaluation method and device based on driving behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109377349A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135973A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | A kind of intelligent credit method based on IM and intelligent credit device |
| CN110667543A (en) * | 2019-10-28 | 2020-01-10 | 吉林大学 | Brake pedal feeling simulation device and method for driving habit classification |
| CN113064883A (en) * | 2020-09-28 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Construction method, computer equipment and storage medium of logistics risk control model |
| CN115186993A (en) * | 2022-06-13 | 2022-10-14 | 中寰卫星导航通信有限公司 | Driver evaluation method, device, equipment and storage medium |
| WO2023054592A1 (en) * | 2021-09-29 | 2023-04-06 | Ｇｌｏｂａｌ Ｍｏｂｉｌｉｔｙ Ｓｅｒｖｉｃｅ株式会社 | Server and program |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140172687A1 (en) * | 2011-08-31 | 2014-06-19 | Mehran Chirehdast | Methods and Systems for Financial Transactions |
| CN106027533A (en) * | 2016-05-25 | 2016-10-12 | 珠海市小源科技有限公司 | Credit investigation method and credit investigation apparatus for mobile terminal |
| CN106095391A (en) * | 2016-05-31 | 2016-11-09 | 携程计算机技术（上海）有限公司 | Based on big data platform and the computational methods of algorithm model and system |
| CN107944772A (en) * | 2017-12-27 | 2018-04-20 | 深圳市轱辘车联数据技术有限公司 | Car based on block chain borrows risk information processing method and processing device |
-
2018
- 2018-09-30 CN CN201811165478.4A patent/CN109377349A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140172687A1 (en) * | 2011-08-31 | 2014-06-19 | Mehran Chirehdast | Methods and Systems for Financial Transactions |
| CN106027533A (en) * | 2016-05-25 | 2016-10-12 | 珠海市小源科技有限公司 | Credit investigation method and credit investigation apparatus for mobile terminal |
| CN106095391A (en) * | 2016-05-31 | 2016-11-09 | 携程计算机技术（上海）有限公司 | Based on big data platform and the computational methods of algorithm model and system |
| CN107944772A (en) * | 2017-12-27 | 2018-04-20 | 深圳市轱辘车联数据技术有限公司 | Car based on block chain borrows risk information processing method and processing device |
Non-Patent Citations (1)
| Title |
|---|
| 李岸杨: ""基于Logistic模型的汽车金融公司个人贷款信用评分研究"", 《中国优秀博硕士学位论文全文数据库（硕士） 经济与管理科学辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135973A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | A kind of intelligent credit method based on IM and intelligent credit device |
| CN110667543A (en) * | 2019-10-28 | 2020-01-10 | 吉林大学 | Brake pedal feeling simulation device and method for driving habit classification |
| CN113064883A (en) * | 2020-09-28 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Construction method, computer equipment and storage medium of logistics risk control model |
| WO2023054592A1 (en) * | 2021-09-29 | 2023-04-06 | Ｇｌｏｂａｌ Ｍｏｂｉｌｉｔｙ Ｓｅｒｖｉｃｅ株式会社 | Server and program |
| JP7565650B2 (en) | 2021-09-29 | 2024-10-11 | Ｇｌｏｂａｌ Ｍｏｂｉｌｉｔｙ Ｓｅｒｖｉｃｅ株式会社 | Server and Program |
| CN115186993A (en) * | 2022-06-13 | 2022-10-14 | 中寰卫星导航通信有限公司 | Driver evaluation method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Campbell et al. | Tone at the bottom: Measuring corporate misconduct risk from the text of employee reviews | |
| US20220172207A1 (en) | Computer-implemented methods for controlling rights related to digital knowledge | |
| CN109377349A (en) | A kind of accrediting amount evaluation method and device based on driving behavior | |
| Alm et al. | New technologies and the evolution of tax compliance | |
| Cotugno et al. | Relationship lending, hierarchical distance and credit tightening: Evidence from the financial crisis | |
| WO2022016102A1 (en) | Systems and methods for controlling rights related to digital knowledge | |
| AU2023314124A1 (en) | Systems and methods for providing process automation and artificial intelligence, market aggregation, and embedded marketplaces for a transactions platform | |
| EP4264533A2 (en) | Market orchestration system for facilitating electronic marketplace transactions | |
| CN113302638A (en) | Method and system for improving machines and systems for automatically performing distributed ledger and other transactions in spot and forward markets for energy, computing, storage, and other resources | |
| US20140032394A1 (en) | Peer to peer lending using a mobile wallet | |
| Flögel et al. | Will FinTech make regional banks superfluous for small firm finance? Observations from soft information‐based lending in Germany | |
| Chye et al. | Credit scoring using data mining techniques | |
| CN107481019A (en) | Order fraud recognition methods, system, storage medium and electronic equipment | |
| CN113989019A (en) | Method, device, equipment and storage medium for identifying risks | |
| Saxena et al. | Big data and Internet of Things (IoT) technologies in Omani banks: A case study | |
| CN108961032A (en) | Borrow or lend money processing method, device and server | |
| CN104321794A (en) | A system and method using multi-dimensional rating to determine an entity's future commercial viability | |
| Zheng et al. | Application of big data technology in insurance innovation | |
| Hoti et al. | Measuring risk in environmental finance | |
| CN109242669A (en) | A kind of credit method, user terminal and server based on trip track | |
| Gourrier | Banking the black community: An analysis of banking among Baltimore’s predominately black communities | |
| CN110210868A (en) | The processing method and electronic equipment of numerical value transfer data | |
| CN117611309A (en) | Abnormal financial transaction identification method and device suitable for father-son authority account | |
| Njoku | Defacement of bank going concern public secrecy | |
| Bourke et al. | Detecting fraudulent financial reporting: teaching the ‘watchdog’new tricks |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190222 |