CN107545356A - Personal social credibility methods of marking based on government data and application - Google Patents
Personal social credibility methods of marking based on government data and application Download PDFInfo
- Publication number
- CN107545356A CN107545356A CN201710516849.8A CN201710516849A CN107545356A CN 107545356 A CN107545356 A CN 107545356A CN 201710516849 A CN201710516849 A CN 201710516849A CN 107545356 A CN107545356 A CN 107545356A
- Authority
- CN
- China
- Prior art keywords
- personal
- evaluation
- credit
- estimate
- value
- 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
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
Present invention is disclosed a kind of personal social credibility methods of marking based on government data and application, personal credit situation is modeled by the use of government affairs big data as initial parameter which employs the method for linear regression and linear programming related data weight is determined by each dimension Segmentation Model, the personal social credibility scoring ultimately formed.Improve the credit evaluation mode in the prior art to user on the market at present and be inclined to personal finance data, can not the personal society's sincerity of quantitatively evaluating completely present situation.
Description
Technical field
The present invention relates to personal social credibility methods of marking.
Background technology
In June, 2014, State Council are printed and distributed on printing and distributing establishment of social credit system planning outline (2014-the year two thousand twenty)
Notice.Outline is pointed out:Operating mechanism is to ensure the institutional basis of each system coordination operation of social credit system.Wherein, keep one's word sharp
Encourage and directly act on each main body of the society behavior of credit, be the core mechanism of social credit system operation.
Mainly there are U.S. FICO scorings for personal credit Rating Model in global range at present:
FICO credit scoring models are the more ripe risk techniques instruments of the credit issuing institutions such as bank application.It is early
In the 1940s, some banks of the U.S. begin to attempt Journal of Sex Research credit-graded approach, for quickly handling a large amount of credits
Application.1956, the famous FICO methods of marking of engineer Bill Fair and mathematician's Earl Isaac joint inventions.Should
Method is that current industry applies most ripe credit scoring model substantially using Logistic homing methods as technological core.
In the 60-80 ages in 20th century, with the progress of information technology and the fast development of business, credit scoring model in credit card, disappear
It is widely applied in expense credit, home mortgage and small business loan.
But the FICO credit scoring services that main flow uses at present, following defect be present, entire population can not be covered, it is special
It is not disadvantaged group;Credit scoring model information dimensional comparison is single, occurs many information dimensions under government affairs big data background,
Data are such as examined and approved, daily life payment data, administrative penalty data, breaks one's promise and is performed data etc., traditional credit score scoring
Effect of the model in terms of government affairs has limitation.Further, since traditional credit evaluation model covering based on FICO scorings
Crowd is narrow, information dimension is single, is lagged on the time.
The content of the invention
The technical problems to be solved by the invention are to realize a kind of broad covered area, evaluation dimension are wide, accuracy is high evaluation
Method.
To achieve these goals, the technical solution adopted by the present invention is:Individual society based on government data and application
Credit-graded approach:
1) obtain and treat that the personal social security of scoring, common reserve fund, vehicle peccancy, telecommunications arrearage, judicial record, personal qualities are commented
Point;
2) respective weights value is multiplied by each scoring of acquisition and obtains social security evaluation of estimate, common reserve fund evaluation of estimate, vehicle peccancy
Evaluation of estimate, telecommunications arrearage evaluation of estimate, judicial records appraisal value, personal qualities evaluation of estimate;
3) add up social security evaluation of estimate, common reserve fund evaluation of estimate, then subtracts telecommunications arrearage evaluation of estimate, judicial records appraisal value, individual
People's quality evaluation of estimate and vehicle peccancy evaluation of estimate obtain evaluation score;
4) evaluation score is added with citizen basis credit score and obtains personal social credibility scoring.(additionally can separately it set
Bonus point item, such as the bonus point standard of good people and good deeds, in step 4) be added in the credit score of civic citizen basis)
The personal social credibility of acquisition is scored compared with default standard credit cut off value, if personal social credibility scores
More than standard credit cut off value, then the individual that the personal social credibility scores belongs to the citizen having a good credit;If personal social credibility
Scoring is less than standard credit cut off value, then the individual of the personal social credibility scoring belongs to the citizen of credit difference.
The social security, common reserve fund, vehicle peccancy, telecommunications arrearage, judicial record, the weighted value of personal qualities be respectively 107,
79th, 2,32,288,134, citizen basis credit is divided into 745.
Present invention employs the method for linear regression and linear programming by the use of government affairs big data as initial parameter to individual
Credit standing modeling determines related data weight by each dimension Segmentation Model, the personal social credibility scoring ultimately formed.Change
The credit evaluation mode in the prior art to user being apt at present on the market is inclined to personal finance data, can not quantify completely
The sincere present situation of the personal society of evaluation.
In addition, the present invention solves the pain spot that government runs into during positive incentive of keeping one's word, for the society of credit main body
Can credit data carry out Quantitative marking grading, by the use of personal credit Rating Model as the reward to main body of keeping one's word and excitation according to
According to the dynamics for determining to commend and reward according to the Score index of quantization.
Embodiment
Personal credit Rating Model based on government data and application, should in order to ensure the accuracy of personal credit point assessment
Pay attention to the scientific principle of mathematical modeling, first, the target group clearly to predict, define the quality of sample populations;Secondly, no
Important data item is included by the number of independent variable quantity, using suitable algorithm;Finally, it is necessary to have sufficiently large
Sample.The difficult point that Logistic is returned is that the weight of explanation index must be estimated with maximum likelihood, using non-linear optimal
Change technology solves.Therefore selection is modeled using the method for linear regression and linear programming to personal credit situation.
By the processing and analysis to big data, show that credit score calculation formula is:
Score=107*I+79*W-2*T-32*M-288*L-134*P+745;
Social security (I), common reserve fund (W), vehicle peccancy (T), telecommunications arrearage (M), judicial record (L), personal qualities (P) are commented
It is divided between 0-1, social security (I), common reserve fund (W), telecommunications arrearage (M) standards of grading are got over according to amount of money height linear evaluation, the amount of money
High decile is higher, and the judicial record (L) of vehicle peccancy (T), according to number, (personal qualities is personal record of bad behavior to personal qualities (P)
Such as steal a ride, the overdue refund of credit card) calculate, standards of grading are according to number height linear evaluation, and number is more, and decile is higher;
Calculated according to big data, the formula fitting degree is higher, and difference scores section is [- 52,69], citizen basis credit score
For 745.
The modeling method of above-mentioned evaluation is according to as follows:
Personal credit Rating Model based on government data and application, should in order to ensure the accuracy of personal credit point assessment
Pay attention to the scientific principle of mathematical modeling, first, the target group clearly to predict, define the quality of sample populations;Secondly,
No matter the number of independent variable quantity is included important data item, using suitable algorithm;Finally, it is necessary to have enough
Big sample.Logistic return difficult point be must with maximum likelihood come estimate explain index weight, using it is non-linear most
Optimisation technique solves.Therefore selection is modeled using the method for linear regression and linear programming to personal credit situation.
Linear regression method
The first step:Sample range (such as civil servant, or common reserve fund pay before several whole city's rankings 25% citizen) is locked, is passed through
Stratified sampling (such as age) chooses 1000 small samples as training set.
Second step:By observing the data (observing more data dimensions, such as credit card repayment record as far as possible) of 1000 people,
Given a mark using expert method for everyone
3rd step:Linear regression is carried out to 1000 people's samples, finds out each key variables and its influence power (power to credit score
Weight)
4th step:Variable is deleted to the further optimization of model by addition
Modeling process example
It can be seen that each variant correlation coefficient is little, substantially conform to variable it is separate it is assumed that correlation highest
Two variables are common reserve fund and telecommunications arrearage (0.4), it may be said that bright wage level is higher, is less susceptible to that telecommunications arrearage occurs.
Most civic credit scores are in average level
Credit score can with social security, common reserve fund increase and increase, with break in traffic rules and regulations, telecommunications arrearage, judicial record and
Personal qualities record increase and reduce.The line of green is more precipitous to illustrate that its influence to credit score is more notable, wherein judicial note
Record is to influence a variable the most significant to credit score.
It is 125 civic score calculations lower to 6 dimensions below:
Call:
Lm (formula=score~insurance+wages+traffic+mobile+legal+personal, data
=newdata)
Residuals:
Min 1Q Median 3Q Max
-52.482 -12.968 -2.193 9.401 69.343
Coefficients:
Estimate Std.Error t value Pr(>|t|)
---
Signif.codes:0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
Residual standard error:23.28 on 118 degrees of freedom Multiple R-
squared:0.9846,Adjusted R-squared:0.9838 F-statistic:1259on 6 and 118 DF,p-
value:<2.2e-16
If the p value of independent variable<0.05, illustrate to may apply in Rating Model credit score with significantly affecting.If p>
0.05, then it is assumed that the independent variable is not notable on credit influence, is added without Rating Model.
Linear programming technique
Assuming that have in sample citizen that ng has a good credit (be labeled as i=1,2,3, ng), the poor citizen of nb credit (mark
I=ng+1, ng+2 are designated as, ng+nb), each citizen have m credit scoring dimension, therefore citizen i characteristic vector is designated as
(xi1, xi2, |, xim)。
In preferable Credit Model, our purpose is to find one group of weight wj(j=1,2, |, m) and one it is critical
Value so that:
For each citizen having a good credit
w1xi1+w2xi2+|wmxim> c
For the citizen of each credit difference
w1xi1+w2xi2+|wmxim< c
C is standard credit cut off value.
Claims (3)
1. the personal social credibility methods of marking based on government data and application, it is characterised in that:
1) obtain and treat the personal social security of scoring, common reserve fund, vehicle peccancy, telecommunications arrearage, judicial record, the scoring of personal qualities;
2) respective weights value is multiplied by each scoring of acquisition and obtains social security evaluation of estimate, common reserve fund evaluation of estimate, vehicle peccancy evaluation
Value, telecommunications arrearage evaluation of estimate, judicial records appraisal value, personal qualities evaluation of estimate;
3) add up social security evaluation of estimate, common reserve fund evaluation of estimate, then subtracts telecommunications arrearage evaluation of estimate, judicial records appraisal value, individual moral standing
Matter evaluation of estimate and vehicle peccancy evaluation of estimate obtain evaluation score;
4) evaluation score is added with citizen basis credit score and obtains personal social credibility scoring.
2. the personal social credibility methods of marking according to claim 1 based on government data and application, it is characterised in that:
The personal social credibility of acquisition is scored compared with default standard credit cut off value, if the scoring of personal social credibility is more than standard
Credit cut off value, the then individual that the personal social credibility scores belong to the citizen having a good credit;If personal social credibility scoring is less than
Standard credit cut off value, the then individual that the personal social credibility scores belong to the citizen of credit difference.
3. the personal social credibility methods of marking according to claim 1 or 2 based on government data and application, its feature exist
In:The social security, common reserve fund, vehicle peccancy, telecommunications arrearage, judicial record, the weighted value of personal qualities are respectively 107,79,2,
32nd, 288,134, citizen basis credit is divided into 745.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710516849.8A CN107545356A (en) | 2017-06-29 | 2017-06-29 | Personal social credibility methods of marking based on government data and application |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710516849.8A CN107545356A (en) | 2017-06-29 | 2017-06-29 | Personal social credibility methods of marking based on government data and application |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107545356A true CN107545356A (en) | 2018-01-05 |
Family
ID=60970952
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710516849.8A Pending CN107545356A (en) | 2017-06-29 | 2017-06-29 | Personal social credibility methods of marking based on government data and application |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107545356A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109254979A (en) * | 2018-09-29 | 2019-01-22 | 中国银行股份有限公司 | A kind of personal credit evaluation method and system |
CN109727016A (en) * | 2018-12-17 | 2019-05-07 | 泰康保险集团股份有限公司 | Credit payment processing method, device, medium and electronic equipment |
CN111694884A (en) * | 2020-06-12 | 2020-09-22 | 广元量知汇科技有限公司 | Intelligent government affair request processing method based on big data |
CN111967790A (en) * | 2020-08-28 | 2020-11-20 | 恒瑞通(福建)信息技术有限公司 | Credit score algorithm model method capable of automatic calculation and terminal |
CN112732812A (en) * | 2020-12-31 | 2021-04-30 | 中国科学技术大学智慧城市研究院(芜湖) | Personal credit analysis method based on big data portrait |
CN112836931A (en) * | 2020-12-31 | 2021-05-25 | 广州智能科技发展有限公司 | Method for generating multidimensional personal credit evaluation model |
-
2017
- 2017-06-29 CN CN201710516849.8A patent/CN107545356A/en active Pending
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109254979A (en) * | 2018-09-29 | 2019-01-22 | 中国银行股份有限公司 | A kind of personal credit evaluation method and system |
CN109727016A (en) * | 2018-12-17 | 2019-05-07 | 泰康保险集团股份有限公司 | Credit payment processing method, device, medium and electronic equipment |
CN111694884A (en) * | 2020-06-12 | 2020-09-22 | 广元量知汇科技有限公司 | Intelligent government affair request processing method based on big data |
CN111967790A (en) * | 2020-08-28 | 2020-11-20 | 恒瑞通(福建)信息技术有限公司 | Credit score algorithm model method capable of automatic calculation and terminal |
CN111967790B (en) * | 2020-08-28 | 2023-04-07 | 恒瑞通(福建)信息技术有限公司 | Credit scoring method capable of automatically calculating and terminal |
CN112732812A (en) * | 2020-12-31 | 2021-04-30 | 中国科学技术大学智慧城市研究院(芜湖) | Personal credit analysis method based on big data portrait |
CN112836931A (en) * | 2020-12-31 | 2021-05-25 | 广州智能科技发展有限公司 | Method for generating multidimensional personal credit evaluation model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107545356A (en) | Personal social credibility methods of marking based on government data and application | |
Anderson | The credit scoring toolkit: theory and practice for retail credit risk management and decision automation | |
Chittedi | Financial crisis and contagion effects to Indian stock market:‘DCC–GARCH’analysis | |
CN110458693A (en) | A kind of automatic measures and procedures for the examination and approval of business loan, device, storage medium and electronic equipment | |
Chamboko et al. | On the modelling of prognosis from delinquency to normal performance on retail consumer loans | |
Chen et al. | The predictive power of Japanese candlestick charting in Chinese stock market | |
CN108492001A (en) | A method of being used for guaranteed loan network risk management | |
CN111340236B (en) | Bond breach prediction method based on bond estimation data and integrated machine learning | |
Liu et al. | CEO overconfidence and bank systemic risk: Evidence from US bank holding companies | |
Bach et al. | Are powerful CEOs beneficial to post-IPO survival in high technology industries?: An empirical investigation | |
CN109242669A (en) | A kind of credit method, user terminal and server based on trip track | |
Voghouei et al. | Does the political institution matter for financial development? | |
Santos et al. | Lobbying on accounting regulation: Evidence from the oil industry | |
Durango et al. | The effects of pricing strategy on the efficiency and self-sustainability of microfinance institutions: a case study | |
CN110533528A (en) | Assess the method and apparatus of business standing | |
Thomsen et al. | Hedge Funds: En Enforcement Perspective | |
CN109801162A (en) | A kind of credit rating method that social media data are merged with multi-standard cross-certification | |
Hatamlou et al. | Forecasting gold price using data mining techniques by considering new factors | |
Varvara et al. | Capital structure in emerging markets: evidence from China | |
Yakubu et al. | Modeling the Effect of Bank Specific Factors on Financial Performance of Commercial Banks in Nigeria: Panel Data Regression Approach | |
Goel et al. | Lending to SMEs: Indian banker's perspective | |
Xie et al. | Analysis of influencing factors of internet lending adoption in China: Internet lending adoption in China | |
Kimondo | The effect of credit policy on the financial performance of deposit taking microfinance institutions in Kenya | |
Botha | A procedure for loss-optimising the timing of loan recovery under uncertainty | |
Rasskazova et al. | Investment simulation model for estimating the future value of tokens |
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: 20180105 |