CN107705036A - Dynamic credit estimation method and system based on multi-dimensional data - Google Patents
Dynamic credit estimation method and system based on multi-dimensional data Download PDFInfo
- Publication number
- CN107705036A CN107705036A CN201711020626.9A CN201711020626A CN107705036A CN 107705036 A CN107705036 A CN 107705036A CN 201711020626 A CN201711020626 A CN 201711020626A CN 107705036 A CN107705036 A CN 107705036A
- Authority
- CN
- China
- Prior art keywords
- user
- social
- association
- payment
- credit
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention proposes a kind of dynamic credit estimation method and system based on multi-dimensional data, including：Obtain social data, the payment data of user；Social strength of association between user is calculated based on social data, the payment strength of association between user is calculated based on payment data, calculates the social payment joint strength of association between user；Determine that the strong association social activity between user pays joint relation and weak rigidity social activity pays joint relation according to the social payment joint strength of association between user；Obtain the credit evaluation value of user；User for not getting credit evaluation value, its credit evaluation value determine according to the social credit evaluation value for paying user associated with it in joint relation of strong association；The multidate information in blacklist storehouse and list storehouse of breaking one's promise is obtained in real time, and corresponding user is updated according to the multidate information in blacklist storehouse and list storehouse of breaking one's promise and there is the credit evaluation value of the social user for paying joint relation of strong association with it.Various dimensions comprehensive assessment is realized, and is assessed more accurately, truly.
Description
Technical field
The present invention relates to financial credit technical field, more particularly to a kind of dynamic credit based on multi-dimensional data to comment
Estimate method and system.
Background technology
Internet finance refers to that conventional banking facilities are real using Internet technology and ICT with Internet enterprises
Existing financing, payment, Novel finical business model (including P2P, the Internet bank, the third party of investment and intermediary information service
Pay etc.).Internet finance includes general favour finance, platform finance, information finance and fragment finance etc. and is different from traditional financial
Financial models.
Internet finance is the useful supplement of existing financial system, and credit system is the foundation stone of modern finance, is being interconnected
Net under financial background, the perfect of credit system is even more the importance for improving internet financial eco-environment, accurately grasps service object
Credit standing, consumption habit and risk partiality it is particularly important.
However, traditional financial industry such as bank will offer loans, it is necessary to carry out credit examination ＆ verification to creditor, emphasis is in kind to be provided
Production, leverage, cash flowing water etc..Meanwhile and due to the hysteresis of China's Credit System Construction, cause substantial amounts of natural person or
The credit record missing of little Wei enterprises, the development to general favour finance bring huge obstacle.
China's credit checking industry is still developing and improved the stage at present, and it develops mainly based on bank, private credit information service
Supplemented by mixing general layout, traditional credit system based on bank is primarily present problems with (defect)：
First, data acquisition channel is relatively simple and fixed during traditional credit evaluation, causes those can not collect money
The user of golden flowing information can not carry out credit evaluation or credit evaluation is not accurate enough, so causes credit evaluation crowd to cover
Rate is largely restricted.
Secondly, current traditional credit system lacks mainly based on the financial transaction of user and financial data information
Various dimensions information, it is only capable of reflecting credit standing of the user in economic aspect, the synthesis credit situation of individual can not be judged.
Furthermore during traditional credit evaluation, the real data renewal of evaluation index and extension it is very slow, evaluated person sees
Credit report be geo-stationary and format, such credit value is true credit situation and the change that can not reflect user
's.
The content of the invention
The technical problems to be solved by the invention be to provide a kind of dynamic credit estimation method based on multi-dimensional data and
System, various dimensions comprehensive assessment is realized, and assessed more accurately, truly.
To solve the above problems, a kind of dynamic credit estimation method based on multi-dimensional data of present invention proposition, including with
Lower step：
S1：Social data, the payment data of user is obtained, the social association calculated based on the social data between user is strong
Degree, the payment strength of association between user is calculated based on the payment data；
S2：Combined according to the social strength of association with the social payment paid between strength of association calculating user and associated by force
Degree；
S3：Determine that social pay of strong association between user is combined according to the social joint strength of association that pays between the user
Relation and the social payment joint relation of weak rigidity；
S4：Obtain the credit evaluation value of user；User for not getting credit evaluation value, its credit evaluation value according to
The social credit evaluation value for paying user associated with it in joint relation of strong association determines；
S5：The multidate information in blacklist storehouse and list storehouse of breaking one's promise is obtained in real time, and according to blacklist storehouse and list storehouse of breaking one's promise
The corresponding user of multidate information renewal and there is the credit evaluation value of the social user for paying joint relation of strong association with it.
According to one embodiment of present invention, in the step S1, it is based respectively on social data and the payment data structure
Build the payment network associated on the social networks of social activity association with payment；Social networks is calculated based on the social data
Social strength of association between middle user, the payment strength of association in payment network between user is calculated based on the payment data.
According to one embodiment of present invention, the social data includes the timely meta of frequency information that social association occurs
Confidence ceases.
According to one embodiment of present invention, the social data includes time location information that social association occurs, most
Number, the number that nearest January, social association occurred of social association generation in nearly 1 year；Social strength of association between the user
Calculation formula is：
β=(γy+γm)*LBS
Wherein, parameter LBS is according to the social time location information determination for associating and occurring, and time and position are closer to then value
It is smaller；Parameter γyDetermine that number more at most value is higher according to the number that association in nearest 1 year occurs；Parameter γmAccording to nearest
The number that January, association occurred determines that number more at most value is higher.
According to one embodiment of present invention, the payment data includes paying frequency information and amount of money letter that association occurs
Breath.
According to one embodiment of present invention, the payment data includes paying within nearest 1 year the number, most that association occurs
Pay the number of association generation nearly January, pay within nearest 1 year the accumulating sum that association occurs；Payment association between the user
Strength calculation formula is：
α=δy+δm+M
Wherein, parameter δyIt is higher according to the number determination for paying association and occurring in nearest 1 year, number more at most value；Parameter
δmThe number determination occurred according to association is paid nearest January, number more at most value are higher；Parameter M was paid according to nearest 1 year
The accumulating sum that association occurs determines that accumulating sum more at most value is higher.
According to one embodiment of present invention, in the step S2, the social payment joint strength of association between the user
Calculation formula be：
θ=β+α
Wherein, β is social strength of association, and α is payment strength of association.
According to one embodiment of present invention, in the step S3, with least each social association user and/or pass is paid
Combination family is that the social joint strength of association θ that pays between node, user is side, builds social activity and pays joint relation map, according to
Social activity payment joint strength of association θ size extracts strong association social activity and paid from the social payment joint relation map to be joined
Conjunction relation and the social payment joint relation of weak rigidity.
According to one embodiment of present invention, the step S4 comprises the following steps：
S41：Obtain each social association user and pay the existing credit evaluation value of association user, and the letter that will be got
Corresponding user is assigned to assessed value；
S42：User for not getting credit evaluation value, its credit evaluation value is entered as to have strong close with the user
The average value of the credit evaluation value of the social user for paying joint relation of connection.
According to one embodiment of present invention, in the step S5, information, renewal bag are fallen into if multidate information is user
Include：
Credit evaluation value zero setting to falling into the user in blacklist storehouse；Pair there is strong pass with falling into the user in blacklist storehouse
The credit rating assessed value of the social user for paying joint relation of connection subtracts certain score value；
Credit evaluation value to the user that falls into list storehouse of breaking one's promise and there is strong association with falling into the user in list storehouse of breaking one's promise
The credit rating assessed value that social activity pays the user of joint relation subtracts certain score value, and the score value that the former subtracts is more than the latter.
According to one embodiment of present invention, in the step S5, information, renewal bag are removed if multidate information is user
Include：To being restored because user corresponding to the user falls into the renewal that information done.
The present invention also provides a kind of dynamic credit estimation method based on multi-dimensional data, including：
Data acquisition module, social data, the payment data for obtaining user are performed, user is calculated based on the social data
Between social strength of association, based on the payment data calculate user between payment strength of association；
Joint strength of association computing module, perform according between the social strength of association and payment strength of association calculating user
Social pay joint strength of association；
Relation decomposing module, perform the strong pass determined according to the social payment joint strength of association between the user between user
Connection is social to pay joint relation and the social payment joint relation of weak rigidity；
Credit evaluation value preliminary design module, perform the credit evaluation value for obtaining user；For not getting credit evaluation value
User, its credit evaluation value are true according to the credit evaluation value of user associated with it in the strong social payment joint relation of association
It is fixed；
Credit evaluation value update module, the real-time multidate information for obtaining blacklist storehouse and list storehouse of breaking one's promise of execution, and according to
The renewal of the multidate information in blacklist storehouse and list storehouse of breaking one's promise corresponds to user and has the social payment joint relation of strong association with it
The credit evaluation value of user.
The present invention also provides a kind of computer-readable recording medium, is stored thereon with computer program, the computer program
When being executed by processor, the dynamic credit evaluation side based on multi-dimensional data as described in any one in previous embodiment is realized
Method.
The present invention also provides a kind of computer equipment, including memory and processor and storage are on a memory and can quilt
The computer program that processor calls, described in the computing device during computer program, realize as any in previous embodiment
The dynamic credit estimation method based on multi-dimensional data described in one.
After adopting the above technical scheme, the present invention has the advantages that compared with prior art：
On the basis of traditional credit evaluation system, social networks and payment network liveness and associated user are introduced
Influence of the credit evaluation to user to be assessed, build inter-related dynamic credit evaluation network between user, this method is effective
The defects of traditional credit estimation method only carries out static evaluation, a weight assets for individual, coverage rate is low is overcome, there is covering
Scope is wide, various dimensions comprehensive assessment, good relevance the advantages that；
On the basis of traditional credit evaluation carries out static data obtained by static credit evaluation to individual, it can also obtain in real time
Dynamic data information influential on credit evaluation, and carried out more according to credit evaluation value of the dynamic data information to relative users
Newly, the credit evaluation drawn is dynamic change, reflects true the credit situation and variation tendency of user in time.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the dynamic credit estimation method based on multi-dimensional data of the embodiment of the present invention；
Fig. 2 is the social networks collection of illustrative plates centered on user A in the embodiment of the present invention；
Fig. 3 is the social networks collection of illustrative plates of 15 users in the embodiment of the present invention；
Fig. 4 is the social payment joint relation map centered on user A in the embodiment of the present invention；
Fig. 5 is that social the social of joint relation that pay of the strong and weak association of band of 15 users in the embodiment of the present invention pays connection
Close relation map；
Fig. 6 adds the schematic diagram for getting existing credit evaluation value on the basis of being Fig. 5；
Completion does not get the schematic diagram of existing credit evaluation value on the basis of Fig. 7 is Fig. 6；
Fig. 8 be Fig. 7 on the basis of based on blacklist storehouse and break one's promise list storehouse be updated after schematic diagram.
Embodiment
In order to facilitate the understanding of the purposes, features and advantages of the present invention, below in conjunction with the accompanying drawings to the present invention
Embodiment be described in detail.
Many details are elaborated in the following description in order to fully understand the present invention.But the present invention can be with
Much it is different from other manner described here to implement, those skilled in the art can be in the situation without prejudice to intension of the present invention
Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.
Referring to Fig. 1, in one embodiment, the dynamic credit estimation method based on multi-dimensional data comprises the following steps：
S1：Social data, the payment data of user is obtained, the social association calculated based on the social data between user is strong
Degree, the payment strength of association between user is calculated based on the payment data；
S2：Combined according to the social strength of association with the social payment paid between strength of association calculating user and associated by force
Degree；
S3：Determine that social pay of strong association between user is combined according to the social joint strength of association that pays between the user
Relation and the social payment joint relation of weak rigidity；
S4：Obtain the credit evaluation value of user；User for not getting credit evaluation value, its credit evaluation value according to
The social credit evaluation value for paying user associated with it in joint relation of strong association determines；
S5：The multidate information in blacklist storehouse and list storehouse of breaking one's promise is obtained in real time, and according to blacklist storehouse and list storehouse of breaking one's promise
The corresponding user of multidate information renewal and there is the credit evaluation value of the social user for paying joint relation of strong association with it.
The dynamic credit estimation method based on multi-dimensional data of the present invention is more particularly described below, but should not
As limit.
In step sl, growing with mobile platform, user terminal can use social platform, payment platform etc.,
The social data and payment data of user can be thus obtained from these social platforms and payment platform.Social platform and branch
It can be same integrated platform or different platform to pay platform, in order that data are more reliable, can take multiple social platforms and
Data on payment platform are counted.
Social strength of association between user is calculated based on the social data, based between payment data calculating user
Pay strength of association.The social degree of strength associated between user and the degree of strength for paying association are calculated respectively, then
Both are joined together as a degree of strength for considering social association and the two-dimentional relation for paying association.
Can be analyzed from the social data of a user user between other users whether it is social associate, Yi Jishe
The power of association is handed over, social networks between these clients and mutual can be obtained by analyzing the social data of each user
Social activity association power；Likewise, can be analyzed from the payment data of a user between the user and other users whether
Pay association and pay the power of association, payment between these clients can be obtained by analyzing the payment data of each user
Network and mutual payment association power.
Social activity association is, for example, to be made a phone call between user and user or social platform engages in the dialogue, circle of friends thumbs up, comments
By etc. situation about both can be associated together；Pay association be, for example, transferred accounts between user and user, receive and dispatch it is red
The situation that both are associated together by bag etc..
Preferably, in step S1, it is based respectively on the social activity that the social data associates with payment data structure on social activity
The payment network that network associates with payment；The social association calculated based on the social data in social networks between user is strong
Degree, the payment strength of association in payment network between user is calculated based on the payment data.
Social networks is the network structure formed using user as the social association between node, user as side；Pay
Network is the network structure formed using user as the payment association between node, user as side.Pass through social networks and branch
Pay network, related user can be quickly found out, so as to when calculating strength of association just for network in annexation be present
User, be easy to calculate.
In one embodiment, positional information between frequency information of the social data including social association generation is timely.It is social
It is closer to associate the contact of the higher explanation of the frequency occurred between the two, if but being sent out in close or similar time, position
When raw, illustrate to be likely to be required situation about frequently occurring between colleague, it is necessary to weaken this influence in credit associates,
Thus the frequency of social association generation is higher, the social association of the more different explanations of time location is stronger.
Preferably, social data includes the time location information of social association generation, social association in nearest 1 year occurs
Number, the number that nearest January, social association occurred, it is contemplated that the time location of generation, and come from long-term contact and short-term contact
Consider, certainly, social data is not limited to this, and the long-term specific time limit for contacting and contacting in short term can also be adjusted, example
Such as nearest 1 year or several years, nearest January or several months.Social strength of association calculation formula between user is：
β=(γy+γm)*LBS
Wherein, parameter LBS is according to the social time location information determination for associating and occurring, and time and position are closer to then value
It is smaller；Parameter γyDetermine that number more at most value is higher according to the number that association in nearest 1 year occurs；Parameter γmAccording to nearest
The number that January, association occurred determines that number more at most value is higher.Can be using each social association user between node, user
Social association degree of strength β be side, structure social networks collection of illustrative plates.
In one embodiment, payment data includes paying frequency information and the amount information that association occurs.Pay association
The frequency of generation is higher, the payment relation of the bigger explanation of the amount of money between the two is closer, thus pays the frequency that association occurs and get over
It is stronger that the bigger explanation of height, the amount of money pays correlation degree.
Preferably, payment data includes the number of payment association generation in nearest 1 year, nearest January pays association generation
Number, the accumulating sum that association occurs is paid within nearest 1 year, equally considered from long-term contact and short-term contact, and synthesis is sent out
The raw amount of money considers the strength of association between two users.Payment strength of association calculation formula between user is：
α=δy+δm+M
Wherein, parameter δyIt is higher according to the number determination for paying association and occurring in nearest 1 year, number more at most value；Parameter
δmThe number determination occurred according to association is paid nearest January, number more at most value are higher；Parameter M was paid according to nearest 1 year
The accumulating sum that association occurs determines that accumulating sum more at most value is higher.Can be using each payment association user as node, use
Payment association degree of strength α between family is side, and structure pays relation map.
Then step S2 is performed, according to the social strength of association and pays the social payment between strength of association calculating user
Joint strength of association.
Preferably, in step S2, the calculation formula of the social payment joint strength of association between the user is：
θ=β+α
Wherein, β is social strength of association, and α can be from two formula in above-described embodiment to pay strength of association
Obtain, but not limited to this, realized social strength of association in the present embodiment and pay strength of association summation and join as social pay
Close strength of association.
Then, step S3 is performed, the strong pass between user is determined according to the social payment joint strength of association between the user
Connection is social to pay joint relation and the social payment joint relation of weak rigidity.
Preferably, in step S3, using at least each social association user and/or association user is paid between node, user
The social joint strength of association θ that pays be side, build it is social pay joint relation map, according to social payment joint strength of association
θ size extracts strong association social activity from the social payment joint relation map and pays joint relation and weak rigidity social activity branch
Pay joint relation.The user of onrelevant can certainly be introduced, subsequently is set to not associate by its strength of association value.
Strong and weak division is carried out the social joint relation that pays user according to preset rules and combination θ value, and looked for
Go out the social payment joint relation of strong association, including corresponding user pays joint strength of association θ, others as weak pass with social
Connection is social to pay joint relation.
Then perform step S4, obtain the credit evaluation value of user, can be the third-party institution or bank made it is each
The credit evaluation value of user, the information that there may be some users certainly fail to collect Capital Flow information and can not carry out letter
With situation about assessing or credit evaluation is not accurate enough, the credit evaluation value missing that the present invention can be to such case carries out polishing；
User for not getting credit evaluation value, its credit evaluation value according to the strong association it is social pay in joint relation with its
The credit evaluation value of the user of association determines.
Preferably, step S4 comprises the following steps：
S41：Obtain each social association user and pay the existing credit evaluation value of association user, and the letter that will be got
Corresponding user is assigned to assessed value；Certainly, can be number by the credit evaluation value pretreatment of nonumericization for the ease of calculating
Handled again after value；
S42：User for not getting credit evaluation value, its credit evaluation value is entered as to have strong close with the user
The average value of the credit evaluation value of the social user for paying joint relation of connection, by paying stronger one of joint relation with its social activity
A little users carry out credit appraisal to it.
Then step S5 is performed, obtains the multidate information in blacklist storehouse and list storehouse of breaking one's promise in real time, and according to blacklist storehouse
There is the strong letter for associating the social user for paying joint relation with the corresponding user of multidate information renewal in list storehouse of breaking one's promise and with it
Use assessed value.
List in blacklist storehouse and list storehouse of breaking one's promise can carry out real-time update according to the situation of reality, have a strong impact on
The meeting of the behavior of credit adds blacklist storehouse, such as the user swindled；The row that charges for water and electricity are not paid or credit card is overdue occurs
Then to add list storehouse of breaking one's promise.If after discreditable behavior amendment or after some behaviors overcome, user can be from blacklist storehouse and mistake
Letter list exits in storehouse.
Preferably, in step S5, if multidate information, which is user, falls into information, that is to say, that there is new user to drop into black
List storehouse or list storehouse of breaking one's promise, then it is corresponding renewal step can include：
Credit evaluation value zero setting to falling into the user in blacklist storehouse；Pair there is strong pass with falling into the user in blacklist storehouse
The credit rating assessed value of the social user for paying joint relation of connection subtracts certain score value；
Credit evaluation value to the user that falls into list storehouse of breaking one's promise and there is strong association with falling into the user in list storehouse of breaking one's promise
The credit rating assessed value that social activity pays the user of joint relation subtracts certain score value, and the score value that the former subtracts is more than the latter.
Specifically, the dynamic blacklist dynamic for obtaining the third-party institution is broken one's promise list, to falling into the use in blacklist storehouse
The credit evaluation value zero setting at family, pair letter for having the user of the social payment joint relation of strong association with falling into the user in blacklist storehouse
Subtract 2 point with assessed value, pair have the credit rating of the social user for paying joint relation of weak rigidity with falling into the user in blacklist storehouse
Assessed value is not deducted points；The credit evaluation value of user to falling into list storehouse of breaking one's promise subtracts 2 point, pair with falling into the use in list storehouse of breaking one's promise
The credit evaluation value that there is the social user for paying joint relation of strong association at family subtracts 1 point, pair with falling into the user in list storehouse of breaking one's promise
The credit rating assessed value for having the social user for paying joint relation of weak rigidity is not deducted points.Certainly, above-mentioned score value without limitation,
It can be adjusted according to being actually needed.
Preferably, in step S5, if multidate information, which is user, removes information, renewal includes：To because corresponding to the user
User falls into the renewal that information is done and restored.
Specifically, the credit evaluation value of the user to removing blacklist storehouse returns to initial credit assessed value, with removal
The credit evaluation value that the user in blacklist storehouse has the social user for paying joint relation of strong association adds 2 points, with removing blacklist storehouse
User have the credit evaluation value of the social user for paying joint relation of weak rigidity not bonus point；User to removing list storehouse of breaking one's promise
Credit evaluation value add 2 points, have the letter of the social user for paying joint relation of strong association with removing the user to break one's promise in list storehouse
With assessed value plus 1 point, the credit for having the social user for paying joint relation of weak rigidity with removing the user in list storehouse of breaking one's promise is commented
Valuation not bonus point.Certainly, above-mentioned score value without limitation, can be adjusted according to being actually needed.
On the basis of traditional credit evaluation, social data and payment data are introduced, adds the dimension of credit evaluation data
Degree so that credit evaluation is more comprehensively more accurate；And using existing credit evaluation value to not getting the user of credit value
Credit evaluation value carry out completion, add credit evaluation population coverage；Meanwhile obtain in real time influential on credit evaluation
Dynamic data information, and be updated according to credit evaluation value of the dynamic data information to relative users, the credit report drawn
It is dynamic change, reflects true the credit situation and variation tendency of user in time.
The embodiment of the present invention is illustrated below by more specifically example.
From the social data got, according to code of points to parameter LBS, γy、γmValue is carried out, three parameters are corresponding
Code of points such as following table (1)：
Table (1)
For example, for user A, getting in its social circle (social connections occur with user A) has B, C, D, E, F five
User, and the time location information of social connections, the nearest 1 year number that social connections occur occurs and occurs nearest January
The number of social connections such as following table (2)：
Table (2)
Based on upper table (2) and β formula are calculated, calculate the degree of strength of social association between user：
βAB=1.3；βAC=0.4；βAD=0.4；βAE=1.1；βAF=0.65；
Wherein, βABSocial strength of association between user A and user B, βACSocial pass between user A and user C
Join intensity, the rest may be inferred.
Using the social association degree of strength β of user A, B, C, D, E, F between node, userAB、βAC、βAD、βAE、βAFFor side,
Social networks collection of illustrative plates is built, as shown in Figure 2.
According to the method described above, user B, C, D, E, F social circle's data are obtained successively, and get and society occurs with user B
The user of crosslinking system has A, H, I, G, O, has A, D, H, I, J, K with the C users that social connections occur, social connections occur with D
User has A, C, E, J, K, L, has A, D, F, L, M, N with the user E users that social connections occur, social connections occur with user F
User have A, E, G, M, N, O, have B, H, F, O with the user G users that social connections occur, social connections occur with user H
User has B, C, G, I, has B, C, H, J with the user I users that social connections occur, has with the user J users that social connections occur
C, D, I, K, with user K occur social connections user have C, D, J, L, with user L occur social connections user have D, E, K,
M, with user M occur social connections user have E, F, N, L, with user N occur social connections user have E, F, O, M, with
The user that social connections occur for family O has G, B, F, N, according to the rule of table (1) the social degree of strength β associated each user
Value is carried out, using user A, B, C, D, E, F, G, H, I, J, K, L, M, N, O as node, the power of the social association between each user
Degree β is that side constructs social networks collection of illustrative plates, as shown in Figure 3 (the degree of strength β for omitting social association).
From the payment data got, according to code of points to parameter δy、δm, M carry out value, comment corresponding to three parameters
Divider is then such as following table (3)：
Table (3)
Continue the example above, for user A, get occur to pay with it contact have tri- users of B, E, F, and most
Pay within nearly 1 year the number occurred, nearest January pays frequency and nearest 1 year pays accumulating sum such as following table (4)：
Table (4)
Based on table (3) and α formula is calculated, calculates and pays the degree of strength of association between user：
αAB=1.1；αAC=0；αAD=0；αAE=0.8；αAF=1.0.
Then, calculate user A and pay strength of association with the social social activity paid between user B, C, D, E, F for associating occurs
θ is respectively：
θAB=βAB+αAB=2.4；
θAC=βAC+αAC=0.4；
θAD=βAD+αAD=0.4；
θAE=βAE+αAE=1.9；
θAF=βAF+αAF=1.65；
Degree of strength θ is associated using each social association user with social pay of the association user between node, user is paid
For side, social payment joint relation map is built.Social pay centered on user A is exemplarily illustrated in the present embodiment to join
Relation map is closed, as shown in figure 4, the social activity of 15 users shown in Fig. 3 pays joint relation map according to can similarly obtain, herein
Omit.
Then, strong and weak division is carried out the social payment relation user according to the value of preset rules and combination θ, preset
Rule can be as follows：
Strong association between two users indicated by the solid line is social to pay joint relation, is represented by dashed line weak between two users
Association is social to pay joint relation, and pays the renewal of joint relation map to the social of 15 users (from user A to user O),
As a result it is as shown in Figure 5.
The user for not getting credit value is carried out by existing credit value in strongly connected social payment joint relation
Completion.Assuming that 15 user A~O in Fig. 5, only 10 have bank or third party's credit evaluation value, under its assessment is available
Table (5) represents：
| Assess source | Individual subject | Credit grade |
| The third-party institution | A | Excellent+ |
| Bank | B | It is excellent |
| Bank | D | Excellent+ |
| The third-party institution | E | Good+ |
| The third-party institution | F | Good+ |
| The third-party institution | G | Excellent+ |
| Bank | I | Excellent+ |
| Bank | J | It is good |
| The third-party institution | L | Difference |
| Bank | N | Difference |
Table (5)
Fig. 5 is updated using above-mentioned credit evaluation value, as shown in Figure 6.
User C, H, K, M, O of credit rating assessment are not provided for the Tu6Zhong third-party institutions or bank, paid by social activity
The credit evaluation value for having the strongly connected social user for paying joint relation with user C, H, K, M, O in joint relation map is led to
Cross in the following manner calculating.
Pair pre-processed with the credit evaluation value that the user has the strongly connected social user for paying joint relation, will
It is digital value that it, which is handled, specific rules such as following table (6)：
| Credit rating is assessed | Reciprocal fraction |
| Excellent+ | 5 |
| It is excellent | 4 |
| Good+ | 3 |
| It is good | 2 |
| Difference | 1 |
| Difference- | 0 |
Then user C, H, K, M, O initial credit assessed value can use in Fig. 6 and those users have strongly connected social activity
The credit evaluation value for paying the user of joint relation takes average calculating：
Such as：
User C in Fig. 6, it will be appreciated from fig. 6 that it is to use to have the strongly connected social user for paying joint relation with user C
Family D, I and J, credit scoring corresponding to user D is 5 points, I is 5 points, J is 2 points, then user C credit evaluation value can then lead to
The fraction for crossing user D, I and J obtains by average, i.e. user C credit evaluation fraction is：It is (5+5+2)/3=4 points, corresponding
Credit rating is evaluated as excellent.
User K in Fig. 6, it will be appreciated from fig. 6 that it is D, J to have the strongly connected social user for paying joint relation with user K
And L, the credit scoring corresponding to the premises are respectively 5 points, 2 points, 1 point, then user K credit rating assess fraction then can be with
Obtained by above-mentioned fraction by average, i.e. the credit rating of user K assesses fraction and is：(5+2+1)/3=2.67 points, fraction warp
After rounding up corresponding credit rating be evaluated as it is good+.
User M in Fig. 6, it will be appreciated from fig. 6 that having with user M points, strongly connected social that pay joint relation is user
N, the credit scoring corresponding to user N are 1 point, then it is 1 point that user M credit rating, which assesses fraction, credit corresponding to the fraction
Degree is evaluated as difference.
User O in Fig. 6, it will be appreciated from fig. 6 that with user O have it is strongly connected it is social pay joint relation be user F and
G, credit scoring corresponding to above-mentioned each user are respectively 5 points, 3 points, then user O credit rating assessment fraction then can be by upper
State fraction to obtain by average, i.e. the credit rating of user O assesses fraction and is：(5+3)/2=4 points, the fraction is after rounding up
Corresponding credit rating is evaluated as excellent.
User H in Fig. 6, it will be appreciated from fig. 6 that with user H have it is strongly connected it is social pay joint relation be user B and
I, the credit scoring corresponding to above-mentioned each user are respectively 5 points, 4 points, then user H credit rating is assessed fraction and can then passed through
Above-mentioned fraction obtains by average, i.e. the credit rating assessment fraction of user H points is：(5+4)/2=4.5 points, the fraction is through four houses
Five enter after corresponding to credit rating be evaluated as it is excellent+.
Result after completion is carried out according to the user to not getting initial credit assessed value, Fig. 6 is updated, such as schemed
Shown in 7.
Then, the multidate information in blacklist storehouse and list storehouse of breaking one's promise in the third-party institution and/or banking institution is obtained in real time,
And the social letter for paying user in joint relation is associated by force to corresponding with the multidate information in list storehouse of breaking one's promise according to blacklist storehouse
It is updated with value.
Blacklist acquisition modes are for example：The dynamic blacklist of the third-party institution is got using extraneous interface, in this implementation
In example, following user is put into dynamic blacklist：
| Numbering | Black list user |
| 1 | J |
| 2 | L |
List storehouse acquisition modes break one's promise for example：The blacklist of bank's discreditable behavior is obtained using extraneous interface, in this implementation
In example, following user has discreditable behavior in recent (in three months)：
According to above-mentioned rule, the social joint relation map that pays described in Fig. 7 is updated, the result after renewal is as schemed
Shown in 8.
Finally, credit ratings of 15 user A~O after being calculated based on the credit estimation method of social payment network can be obtained
Assessed value such as following table (7)：
Table (7)
The present invention also provides a kind of dynamic credit evaluation system based on multi-dimensional data, including：
Data acquisition module, social data, the payment data for obtaining user are performed, user is calculated based on the social data
Between social strength of association, based on the payment data calculate user between payment strength of association；
Joint strength of association computing module, perform according between the social strength of association and payment strength of association calculating user
Social pay joint strength of association；
Relation decomposing module, perform the strong pass determined according to the social payment joint strength of association between the user between user
Connection is social to pay joint relation and the social payment joint relation of weak rigidity；
Credit evaluation value preliminary design module, perform the credit evaluation value for obtaining user；For not getting credit evaluation value
User, its credit evaluation value are true according to the credit evaluation value of user associated with it in the strong social payment joint relation of association
It is fixed；
Credit evaluation value update module, the real-time multidate information for obtaining blacklist storehouse and list storehouse of breaking one's promise of execution, and according to
The renewal of the multidate information in blacklist storehouse and list storehouse of breaking one's promise corresponds to user and has the social payment joint relation of strong association with it
The credit evaluation value of user.
The present invention also provides a kind of computer-readable recording medium, is stored thereon with computer program, the computer program
When being executed by processor, the dynamic credit evaluation side based on multi-dimensional data as described in any one in previous embodiment is realized
Method.
The present invention also provides a kind of computer equipment, including memory and processor and storage are on a memory and can quilt
The computer program that processor calls, described in the computing device during computer program, realize as any in previous embodiment
The dynamic credit estimation method based on multi-dimensional data described in one.
Dynamic credit evaluation system, computer-readable recording medium and the calculating based on multi-dimensional data on the present invention
The particular content of machine equipment may refer to the tool in the dynamic credit estimation method based on multi-dimensional data of previous embodiment
Body description content, will not be repeated here.
Although the present invention is disclosed as above with preferred embodiment, it is not for limiting claim, any this area
Technical staff without departing from the spirit and scope of the present invention, can make possible variation and modification, therefore the present invention
Protection domain should be defined by the scope that the claims in the present invention are defined.
Claims (14)
1. a kind of dynamic credit estimation method based on multi-dimensional data, it is characterised in that comprise the following steps：
S1：Social data, the payment data of user is obtained, the social strength of association between user is calculated based on the social data,
Payment strength of association between user is calculated based on the payment data；
S2：According to the social strength of association and pay the social payment joint strength of association between strength of association calculating user；
S3：Determine that the strong association social activity between user pays joint relation according to the social joint strength of association that pays between the user
With the social payment joint relation of weak rigidity；
S4：Obtain the credit evaluation value of user；User for not getting credit evaluation value, its credit evaluation value is according to
The social credit evaluation value for paying user associated with it in joint relation of strong association determines；
S5：Obtain blacklist storehouse and the multidate information in list storehouse of breaking one's promise in real time, and according to the dynamic of blacklist storehouse and list storehouse of breaking one's promise
State information updating corresponds to user and has the credit evaluation value of the social user for paying joint relation of strong association with it.
2. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the step S1
In, it is based respectively on the branch that the social data associates with payment data structure on the social networks that social activity associates with payment
Pay network；Social strength of association in social networks between user is calculated based on the social data, based on the payment data meter
Calculate the payment strength of association between user in payment network.
3. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the social number
According to the frequency information occurred including social association it is timely between positional information.
4. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 3, it is characterised in that the social number
The number occurred, social association in nearest January are associated according to the time location information occurred including social association, social activity in nearest 1 year
The number of generation；Social strength of association calculation formula between the user is：
β=(γy+γm)*LBS
Wherein, parameter LBS according to the social time location information determination for associating and occurring, closer to then value get over by time and position
It is small；Parameter γyDetermine that number more at most value is higher according to the number that association in nearest 1 year occurs；Parameter γmAccording to nearest one
The number that moon association occurs determines that number more at most value is higher.
5. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the payment number
According to including paying frequency information and the amount information that association occurs.
6. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 5, it is characterised in that the payment number
The number occurred, the number that nearest January, payment association occurred, payment association in nearest 1 year are associated according to being paid including nearest 1 year
The accumulating sum of generation；Payment strength of association calculation formula between the user is：
α=δy+δm+M
Wherein, parameter δyIt is higher according to the number determination for paying association and occurring in nearest 1 year, number more at most value；Parameter δmRoot
The number determination occurred according to association is paid nearest January, number more at most value are higher；Parameter M paid according to nearest 1 year and associated
The accumulating sum of generation determines that accumulating sum more at most value is higher.
7. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the step S2
In, the calculation formula of the social payment joint strength of association between the user is：
θ=β+α
Wherein, β is social strength of association, and α is payment strength of association.
8. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 7, it is characterised in that the step S3
In, using at least each social association user and/or pay social payment joint strength of association of the association user between node, user
θ is side, and structure is social to pay joint relation map, is paid according to the social size for paying joint strength of association θ from the social activity
The social payment joint relation of strong association is extracted in joint relation map and weak rigidity social activity pays joint relation.
9. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the step S4
Comprise the following steps：
S41：The existing credit evaluation value of each user is obtained, and the credit evaluation value got is assigned to corresponding user；
S42：User for not getting credit evaluation value, its credit evaluation value is entered as having with the user by force associates society
Hand over the average value of the credit evaluation value for the user for paying joint relation.
10. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 1, it is characterised in that the step
In S5, if multidate information, which is user, falls into information, renewal includes：
Credit evaluation value zero setting to falling into the user in blacklist storehouse；Pair have with falling into the user in blacklist storehouse and to associate society by force
The credit rating assessed value for the user for paying joint relation is handed over to subtract certain score value；
Credit evaluation value to the user that falls into list storehouse of breaking one's promise and have with falling into the user in list storehouse of breaking one's promise and associate social activity by force
The credit rating assessed value for paying the user of joint relation subtracts certain score value, and the score value that the former subtracts is more than the latter.
11. the dynamic credit estimation method based on multi-dimensional data as claimed in claim 10, it is characterised in that the step
In S5, if multidate information, which is user, removes information, renewal includes：To being done more because user corresponding to the user falls into information
Newly restored.
A kind of 12. dynamic credit evaluation system based on multi-dimensional data, it is characterised in that including：
Data acquisition module, social data, the payment data for obtaining user are performed, based between social data calculating user
Social strength of association, the payment association calculated based on the payment data between user are strong；
Joint strength of association computing module, perform according to the social strength of association and pay the society between strength of association calculating user
Hand over and pay joint strength of association；
Relation decomposing module, perform the strong association society determined according to the social payment joint strength of association between the user between user
Hand over and pay joint relation and the social payment joint relation of weak rigidity；
Credit evaluation value preliminary design module, perform the credit evaluation value for obtaining user；User for not getting credit evaluation value,
Its credit evaluation value determines according to the social credit evaluation value for paying user associated with it in joint relation of the strong association；
Credit evaluation value update module, the multidate information for obtaining blacklist storehouse and list storehouse of breaking one's promise in real time is performed, and according to black name
The renewal of the multidate information in single storehouse and list storehouse of breaking one's promise corresponds to user and has the user of the social payment joint relation of strong association with it
Credit evaluation value.
13. a kind of computer-readable recording medium, is stored thereon with computer program, the computer program is executed by processor
When, realize the dynamic credit estimation method based on multi-dimensional data as described in any one in claim 1-11.
14. a kind of computer equipment, including memory and processor and storage are on a memory and can be by processor calling
Computer program, described in the computing device during computer program, realize as described in any one in claim 1-11
Dynamic credit estimation method based on multi-dimensional data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711020626.9A CN107705036A (en) | 2017-10-27 | 2017-10-27 | Dynamic credit estimation method and system based on multi-dimensional data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711020626.9A CN107705036A (en) | 2017-10-27 | 2017-10-27 | Dynamic credit estimation method and system based on multi-dimensional data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107705036A true CN107705036A (en) | 2018-02-16 |
Family
ID=61181556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711020626.9A Pending CN107705036A (en) | 2017-10-27 | 2017-10-27 | Dynamic credit estimation method and system based on multi-dimensional data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107705036A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564467A (en) * | 2018-05-09 | 2018-09-21 | 平安普惠企业管理有限公司 | A kind of determination method and apparatus of consumer's risk grade |
| CN108615119A (en) * | 2018-05-09 | 2018-10-02 | 平安普惠企业管理有限公司 | A kind of recognition methods of abnormal user and equipment |
| CN108647301A (en) * | 2018-05-09 | 2018-10-12 | 平安普惠企业管理有限公司 | A kind of creation method and terminal device of customer relationship net |
| CN108846743A (en) * | 2018-06-12 | 2018-11-20 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
| CN109087163A (en) * | 2018-07-06 | 2018-12-25 | 阿里巴巴集团控股有限公司 | The method and device of credit evaluation |
| CN109670944A (en) * | 2018-12-19 | 2019-04-23 | 信雅达系统工程股份有限公司 | A kind of rating business credit method and system based on map relational network |
| CN110009487A (en) * | 2019-04-10 | 2019-07-12 | 湖北风口网络科技有限公司 | A kind of finance debt-credit methods of risk assessment and system |
| CN110020940A (en) * | 2019-04-02 | 2019-07-16 | 中电科大数据研究院有限公司 | Processing method, device, equipment and the storage medium of credit list |
| CN110633804A (en) * | 2019-09-12 | 2019-12-31 | 深圳众赢维融科技有限公司 | Financial product incoming item evaluation method, server and storage medium |
| CN111242656A (en) * | 2018-11-28 | 2020-06-05 | 中国移动通信集团浙江有限公司 | User credit evaluation method and system based on telecommunication big data |
| CN112041881A (en) * | 2019-04-26 | 2020-12-04 | 天空株式会社 | Server for evaluating user through analysis of social network, method and system for evaluating user through analysis of social network |
| CN113285806A (en) * | 2021-05-10 | 2021-08-20 | 湖南大学 | Dynamic execution method and system for control instruction of power heterogeneous equipment |
-
2017
- 2017-10-27 CN CN201711020626.9A patent/CN107705036A/en active Pending
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108615119B (en) * | 2018-05-09 | 2024-02-06 | 广州地铁小额贷款有限公司 | Abnormal user identification method and equipment |
| CN108615119A (en) * | 2018-05-09 | 2018-10-02 | 平安普惠企业管理有限公司 | A kind of recognition methods of abnormal user and equipment |
| CN108647301A (en) * | 2018-05-09 | 2018-10-12 | 平安普惠企业管理有限公司 | A kind of creation method and terminal device of customer relationship net |
| CN108564467A (en) * | 2018-05-09 | 2018-09-21 | 平安普惠企业管理有限公司 | A kind of determination method and apparatus of consumer's risk grade |
| CN108846743A (en) * | 2018-06-12 | 2018-11-20 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
| CN109087163A (en) * | 2018-07-06 | 2018-12-25 | 阿里巴巴集团控股有限公司 | The method and device of credit evaluation |
| CN109087163B (en) * | 2018-07-06 | 2021-07-09 | 创新先进技术有限公司 | Credit assessment method and device |
| CN111242656A (en) * | 2018-11-28 | 2020-06-05 | 中国移动通信集团浙江有限公司 | User credit evaluation method and system based on telecommunication big data |
| CN109670944A (en) * | 2018-12-19 | 2019-04-23 | 信雅达系统工程股份有限公司 | A kind of rating business credit method and system based on map relational network |
| CN110020940A (en) * | 2019-04-02 | 2019-07-16 | 中电科大数据研究院有限公司 | Processing method, device, equipment and the storage medium of credit list |
| CN110009487A (en) * | 2019-04-10 | 2019-07-12 | 湖北风口网络科技有限公司 | A kind of finance debt-credit methods of risk assessment and system |
| CN112041881A (en) * | 2019-04-26 | 2020-12-04 | 天空株式会社 | Server for evaluating user through analysis of social network, method and system for evaluating user through analysis of social network |
| CN110633804A (en) * | 2019-09-12 | 2019-12-31 | 深圳众赢维融科技有限公司 | Financial product incoming item evaluation method, server and storage medium |
| CN110633804B (en) * | 2019-09-12 | 2022-06-24 | 简链科技(广东)有限公司 | Financial product incoming evaluation method, server and storage medium |
| CN113285806B (en) * | 2021-05-10 | 2022-04-29 | 湖南大学 | Dynamic execution method and system for control instruction of power heterogeneous equipment |
| CN113285806A (en) * | 2021-05-10 | 2021-08-20 | 湖南大学 | Dynamic execution method and system for control instruction of power heterogeneous equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107705036A (en) | Dynamic credit estimation method and system based on multi-dimensional data | |
| CN108665159A (en) | A kind of methods of risk assessment, device, terminal device and storage medium | |
| Gong et al. | Home country investor protection, ownership structure and cross‐listed firms' compliance with SOX‐Mandated internal control deficiency disclosures | |
| CN109829776A (en) | Trade company's methods of risk assessment, device, computer equipment and storage medium | |
| CN111192131A (en) | Financial risk prediction method and device and electronic equipment | |
| Koppell | Political control for China's state‐owned enterprises: lessons from America's experience with hybrid organizations | |
| CN110866822B (en) | Wind control management method and device for securitization of assets, electronic equipment and storage medium | |
| CN104992234A (en) | Enterprise risk assessment method based on various kinds of operation data | |
| CN109740914A (en) | A kind of method, storage medium, equipment and system that financial business is assessed, recommended | |
| CN109345372A (en) | Credit-graded approach, system and computer readable storage medium | |
| Yun et al. | Capital structure optimization for build–operate–transfer (BOT) projects using a stochastic and multi-objective approach | |
| CN110533521A (en) | Method for early warning, device, equipment and readable storage medium storing program for executing after dynamic is borrowed | |
| CN110472806A (en) | Financial letter comments System and method for | |
| CN107909516A (en) | A kind of problem source of houses recognition methods and system | |
| CN107491865A (en) | The social credibility appraisal procedure and device of a kind of user | |
| Levine et al. | Dodd-Frank is a Pigouvian regulation | |
| CN109308667A (en) | A kind of investment risk Prediction System based on big data | |
| CN117350588A (en) | Comprehensive evaluation method and evaluation model for skill ability of financial department | |
| CN109727116A (en) | Credit analysis method, device, equipment and computer readable storage medium | |
| CN111798304A (en) | Risk loan determination method and device, electronic equipment and storage medium | |
| CN111160695A (en) | Method, system, device and storage medium for identifying risk account of computer operation | |
| CN111047092A (en) | Dispute case victory rate prediction method and device, computer equipment and storage medium | |
| Kumar | Linking remittances with financial development and ICT: a study of the Philippines | |
| CN113327161A (en) | Intelligent wind control decision system for credit business | |
| CN113888278A (en) | Data analysis method and device based on enterprise credit line analysis model |
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: 20180216 |