CN108615191A - A kind of credit line intelligent evaluation method - Google Patents
A kind of credit line intelligent evaluation method Download PDFInfo
- Publication number
- CN108615191A CN108615191A CN201810415598.9A CN201810415598A CN108615191A CN 108615191 A CN108615191 A CN 108615191A CN 201810415598 A CN201810415598 A CN 201810415598A CN 108615191 A CN108615191 A CN 108615191A
- Authority
- CN
- China
- Prior art keywords
- credit
- data
- value
- line
- probability
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Business, Economics & Management (AREA)
- Technology Law (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Algebra (AREA)
- Development Economics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention discloses a kind of credit line intelligent evaluation method, and step includes：S1., user is normally fulfiled to the probability of happening of agreement as initial credit score value, and using the model of user credit data training probability, obtains initial credit Rating Model；S2. the credit data for obtaining user to be assessed scores to the credit data of acquisition using initial credit Rating Model, obtains initial credit score value；S3. initial credit score value is input in the probability distribution relationship between the credit scoring value built in advance according to probability density characteristics and line of credit angle value, the line of credit angle value of user to be assessed is calculated.The present invention has many advantages, such as the intelligent evaluation that implementation method is simple, can realize credit line, and assesses efficiency and precision height, using flexible.
Description
Technical field
The present invention relates to credit evaluation technical field more particularly to a kind of credit line intelligent evaluation methods.
Background technology
Credit rating refers to paying one's debts this in full amount as scheduled to debtor by independent third party's credit rating intermediary
The ability and wish of breath are evaluated, and the severity of simple grading symbolic indication its default risk and loss, broad sense is used in combination
Credit rating refer to then that the overall assessment of related contract and the economic ability and wish promised to undertake is fulfiled to grading object.Credit machine
When structure receives customers' credit application, the characteristic variable in the application form submitted using client is established Rating Model and obtains applicant's
One credit value judges the overdue possibility of the borrower, to decide whether by the value compared with the standard value being previously set
Credit and the accrediting amount are granted, such credit scoring is to apply for scoring.The methods of marking Main Basiss of application scoring are client
Personal information is broadly divided into four parts：First, personal essential information, the mainly name including client, working condition, inhabitation
Address, education degree etc.；Second is that personal transaction record, mainly a situation arises for the business of client and financial institution；Third, visitor
The personal credit history at family, it is mainly personal from financial institution loan situation, situation of repaying the loan etc.；Fourth, open record feelings
The open judgement or bankruptcy situation etc. of condition, mainly law court about client.After obtaining personal credit information, credit mechanism is logical
It crosses and establishes personal credit Rating Model and obtain the credit scoring of client, credit scoring shows the corresponding credit grade of client, and
Credit mechanism then gives the different accrediting amount of client according to this credit scoring.
It is assessed for credit line, is all usually at present that the maximum simply born according to historical experience and enterprise is put
Amount is borrowed, a set of amount is established and corresponds to table, be simple linear distribution characteristic between scoring and amount, i.e., as scoring is high
It is low, correspond to corresponding lending amount according to a certain percentage.Such credit line assessment mode is realized simply, but assessment performance
Difference, and flexibility is not strong, when amount is changed, needs to readjust the entire corresponding table of amount scoring, if amount needs frequency
Numerous adjustment can then greatly increase the complexity that assessment is realized, reduce assessment efficiency.
To solve the above problems, have practitioner propose using data digging method realize credit line assess, i.e., according to
The credit amount that past debt-credit data are marked, it is final according to quasi- using complicated machine learning algorithm into the fitting of line function
The function of conjunction carries out amount assessment.But such method realizes complexity, needs using complicated machine learning algorithm, such as nerve
Network algorithm realization is extremely complex, and a large amount of data is needed to provide training, after updating the data every time, is required for regenerating
Model, data processing amount is big, and the tuning of parameter is also very cumbersome, and the efficiency of entire evaluation process is low, of high cost, for real-time
It is required that high occasion, such method is simultaneously not suitable for.
Invention content
The technical problem to be solved in the present invention is that：For technical problem of the existing technology, the present invention provides one
Kind implementation method is simple, can realize the intelligent evaluation of credit line, and assesses efficiency and precision height, using flexible line of credit
Spend intelligent evaluation method.
In order to solve the above technical problems, technical solution proposed by the present invention is：
A kind of credit line intelligent evaluation method, step include：
S1., user is normally fulfiled to the probability of happening p of agreement as initial credit score value, and uses user credit data
The model of training Probability p, obtains initial credit Rating Model；
S2. the credit data for obtaining user to be assessed, using the initial credit Rating Model to the credit data of acquisition
It scores, obtains initial credit score value；
S3., the initial credit score value is input to the credit scoring value built in advance according to probability density characteristics and letter
With the line of credit angle value that user to be assessed in the distribution relation between amount value, is calculated.
As a further improvement on the present invention：The mould of Logic Regression Models training Probability p is specifically based in the step S1
Type obtains initial credit Rating Model.
As a further improvement on the present invention, the step S1 the specific steps are：
S11. it obtains original user credit data and carries out data prediction, data after output pretreatment；
S12. discretization data in data after the pretreatment are subjected to numeralization processing, with use numerical identity items from
The attribute of dispersion data, obtains numeric type data；
S13. logic-based regression model is trained the obtained numeric type data, obtains logic-based and returns mould
The initial credit Rating Model of type.
As a further improvement on the present invention：When carrying out data prediction in the step S11, specifically search described original
Missing values, redundancy value and exceptional value in user credit data, and the missing values found are subjected to Lagrange's interpolation
Processing, and the redundancy value found, exceptional value are subjected to delete processing.
As a further improvement on the present invention：Discretization data carry out numerical value in data after the step S12 will be pre-processed
When changing processing, the data with incremental relationship type are specifically used into incremental numerical identity, without the data difference for being incremented by relationship type
Multiple attributes are added, each attribute is identified using numerical value.
As a further improvement on the present invention, when the logic-based regression model builds the credit scoring model, tool
Body first does a logical transition to p：
A logical transition first specifically is done to p：
L=β0+β1x1+βixi++βnxn=βTx
Wherein, i represents each attribute, and β is the weight coefficient of each attribute, xiFor the value under the attribute, β=(β0,β1,,
βi,βn)T, x=(x0,x1,,xi,xn)T；
The function model that final structure obtains p is：
The function model of the p obtained by above-mentioned structure is compressed to predicted value is exported between [0,1].
As a further improvement on the present invention, the step S2 the specific steps are：
S21. the original credit data for obtaining user to be assessed carries out data prediction, data after output pretreatment；
S22. data after pretreatment that the step S21 is obtained are subjected to numeralization processing, to use numerical identity every
The attribute of discretization data, obtains numeric type data；
S23. the obtained numeric type datas of the step S22 are input in the initial credit Rating Model, are obtained just
Beginning credit scoring value.
As a further improvement on the present invention, the probability distribution relationship between the credit scoring value and line of credit angle value is
Normal distribution or t distributions, and meet：
Wherein, P is credit scoring value, and Q is line of credit angle value, and f (Q) is probability-distribution function.
As a further improvement on the present invention, further include pre-establishing credit scoring value and credit line in the step S3
Mapping table between value calculates to after the line of credit angle value, corresponding credit is found out from the mapping table
Score value obtains the final credit scoring value output of user to be assessed.
Compared with the prior art, the advantages of the present invention are as follows：
1) credit line intelligent evaluation method of the present invention is first building user just using the probability density characteristics of lend-borrow action
The model for often fulfiling the probability of happening of agreement carries out initial score using the model of structure to user, and obtained initial credit is commented
Score value is probability value, and corresponding line of credit angle value can be calculated by being based on distribution character by the probability value, and appraisal procedure is simple, nothing
Complicated calculating process is needed, quick credit line assessment may be implemented, and take full advantage of the credit line point of lend-borrow action
Cloth characteristic indicates the relationship between amount and scoring using simple linear model compared to tradition, can effectively improve amount
Assessment performance, and application is flexible.
2) credit line intelligent evaluation method of the present invention further utilizes the probability density characteristics of lend-borrow action, by credit
Distribution relation between score value and line of credit angle value is equivalent to certain probability-distribution function, can accurate characterization credit scoring with
Distribution relation between credit line, to realize accurate credit line assessment, and can be in maximum based on probability density characteristics
Bear adjust automatically parameter within the scope of lending.
3) credit line intelligent evaluation method of the present invention is based further on Logic Regression Models structure credit scoring model,
Model output value is the probability value of numeric type, thus correspondence, base can be just formed with the probability-distribution function of credit line
It also maps within arbitrary required scoring section in probability value, while can be well solved using Logic Regression Models
Nonlinear problem between dependent variable and explanatory variable, and the speed to score is fast, efficient, after the completion of algorithm training, directly makes
Credit scoring can be carried out to new user, can further increase the precision and effect of credit scoring with the weight parameter of generation
Rate.
Description of the drawings
Fig. 1 is the implementation process schematic diagram of the present embodiment credit line intelligent evaluation method.
Fig. 2 is the probability-distribution function distribution curve schematic diagram of credit line in the present embodiment.
Fig. 3 is the implementation process schematic diagram that initial credit score value is calculated in the present embodiment.
Specific implementation mode
Below in conjunction with Figure of description and specific preferred embodiment, the invention will be further described, but not therefore and
It limits the scope of the invention.
As shown in Figure 1, the present embodiment credit line intelligent evaluation method, step include：
S1., user is normally fulfiled to the probability of happening p of agreement as initial credit score value, and uses user credit data
The model of training Probability p, obtains initial credit Rating Model；
S2. the credit data for obtaining user to be assessed carries out the credit data of acquisition using initial credit Rating Model
Scoring, obtains initial credit score value；
S3., initial credit score value is input to the credit scoring value and line of credit built in advance according to probability density characteristics
In distribution relation between angle value, the line of credit angle value of user to be assessed is calculated.
It is independent of each other between the lend-borrow action data of each user, there is certain probability density characteristics, this implementation
Example utilizes the probability density characteristics of lend-borrow action, first builds the model that user normally fulfils the probability of happening p of agreement, uses structure
Model initial score is carried out to user, obtained initial credit score value is probability value, and it is special to be based on distribution by the probability value
Property can calculate corresponding line of credit angle value, appraisal procedure is simple, and without complicated calculating process, quick credit may be implemented
Amount is assessed, and takes full advantage of the credit line distribution character of lend-borrow action, and simple linear model is used compared to tradition
It indicates the relationship between amount and scoring, can effectively improve amount assessment performance and application is flexible.
Distribution relation between the credit scoring value built in the present embodiment and line of credit angle value is certain probability distribution.If
A certain event generation is influenced by many mutually independent enchancement factors, when influencing very small caused by each factor, always
Influence be considered in all respects as obeying corresponding probability distribution relationship, and for the lend-borrow action of user, the debt-credit row of each user
Be independent of each other between data, and between the influence that generates it is very small, you can probability distribution relationship should be obeyed, such as just
State distribution, Student distributions etc., as described in Figure 2, the probability density function f (Q) of distribution indicate that Q occurs under certain conditions
Probability value, this probability value is within [0,1] section；The probability of happening that user breaks a contract is denoted as 0 by the present embodiment, is normally kept one's word and is abided by
The loaning bill behavior probability for following treaty fixed is denoted as 1, wherein 1 probability occurred is denoted as p, p ∈ (0,1) indicate that user to be assessed normally carries out
The probability of happening of row agreement is p, and the probability that user's promise breaking occurs is 1-p, the interval range of probability density function f (Q) and Probability p
Unanimously.
The present embodiment utilizes the probability density characteristics of lend-borrow action, can be between accurate characterization credit scoring and credit line
Distribution relation, to realize the assessment of accurate credit line, and lending range can be born in maximum based on probability density characteristics
Interior adjust automatically parameter.
In the present embodiment, it is specifically based on the model of logic (Logistic) regression model training Probability p in step S1, obtains
Initial credit Rating Model.Logistic regression analyses are a kind of recurrence point solving nonlinear problem with linear regression model (LRM)
Analysis method, the present embodiment logic-based regression model build credit scoring model, and model output value is the probability value of numeric type, because
And correspondence can be just formed with the probability-distribution function of credit line, it is also mapped to based on probability value arbitrary required
Within scoring section, while Logic Regression Models directly react the credit of user using the Default Probability of borrower as dependent variable
Value will filter out with the relevant ATTRIBUTE INDEX of user credit as explanatory variable, can well solve dependent variable and explain
Nonlinear problem between variable, and the speed to score is fast, efficient, after the completion of algorithm training, directly using the weight generated
Parameter can carry out credit scoring to new user, can further increase the precision and efficiency of credit scoring.
As shown in figure 3, in the present embodiment step S1 the specific steps are：
S11. it obtains original user credit data and carries out data prediction, data after output pretreatment；
S12. discretization data in data after pretreatment are subjected to numeralization processing, to use numerical identity items discretization
The attribute of data, obtains numeric type data；
S13. logic-based regression model is trained obtained numeric type data, obtains logic-based regression model
Initial credit Rating Model.
When carrying out data prediction in the present embodiment, in step S11, it is specific search original user credit data (including with
All kinds of credit datas such as essential information, the transaction record information at family) in missing values (i.e. vacancy value), redundancy value and exceptional value,
And the missing values found are subjected to Lagrange's interpolation processing, the redundancy value found, exceptional value are subjected to delete processing.I.e.
Data prediction includes redundancy value processing：The data repeated or invalid data are deleted；Missing values are handled：It will lack
Mistake value (vacancy value) carries out the bright difference of controlling of glug and handles；Outlier processing：It is more than preset range or bright by the attribute of some data
The aobvious data more than cognitive range are deleted, such as the user that the age is 150 years old.
Since the input data of Logic Regression Models need to be the data of numeric type, after the present embodiment step S12 will be pre-processed
When discretization data carry out numeralization processing in data, the data with incremental relationship type are specifically used into incremental numerical value mark
Know, such as academic type data, the primary school of user, junior middle school, senior middle school and university's attribute can be waited mark using incremental value 0,1,2,3
Know；Multiple attributes are added respectively without the data for being incremented by relationship type, and each attribute is identified using numerical value, such as video, the registered permanent residence, electricity
Words, identity card, certification type etc. are added to multiple attributes respectively, and each attribute-bit is 0,1, that is, uses (0,1) to identify whether
Video verification uses (0,1) to identify whether registered permanent residence verification etc..Logic Regression Models can be used to carry out after the completion of numeralization processing
Training, to obtain logistic regression credit scoring model.
In the present embodiment, when logic-based regression model builds the credit scoring model, a logic first specifically is done to p
Conversion：
L=β0+β1x1+βixi++βnxn=βTx (1)
Wherein, i represents each attribute, and β is the weight coefficient of each attribute, xiFor the value under the attribute, β=(β0,β1,,
βi,βn)T, x=(x0,x1,,xi,xn)T；
The function model that final structure obtains p is：
The function model of the p obtained by above-mentioned structure is compressed to predicted value is exported between [0,1].
Since the probability of happening that borrower normally fulfils agreement is p, then the probability that borrower's promise breaking occurs is 1-p, then
The Probability p that borrower normally fulfils agreement just becomes the object studied in model, and Logistic regression analyses are not directly to general
Rate p establishes model, but does a logical transition to p, converts as described in above-mentioned formula (1), by formula (1) can be seen that L ∈ (- ∞,
+ ∞) andTherefore the linear regression as shown in above formula (2) can be carried out to L, final structure obtains the function of p
Shown in for example above-mentioned formula (3) of model, which can effectively ensure p ∈ (0,1), and the credit value p of borrower is with set of variables
The variation of conjunction value and consecutive variations, the credit rating that p tends to 1 expression borrower are higher, it is on the contrary then the Default Probability of borrower is higher.
In the present embodiment, step S2 the specific steps are：
S21. the original credit data for obtaining user to be assessed carries out data prediction, data after output pretreatment；
S22. data carry out numeralization processing after pretreatment step S21 obtained, to use numerical identity items discrete
The attribute for changing data, obtains numeric type data；
S23. the obtained numeric type datas of step S22 are input in initial credit Rating Model, obtain initial credit and comments
Score value.
When needing to carry out credit line assessment to new user, the original credit data for obtaining user to be assessed carries out data
Pretreatment, line number of going forward side by side value processing, data prediction and numeralization processing method as detailed above, numeralization processing after
The numeric type data needed for Logic Regression Models is obtained, enters data into the initial credit Rating Model built in advance and carries out
It calculates, by model output value as initial credit score value, which corresponds to the probability value of [0,1].
Further include the size for judging initial credit score value after the present embodiment step S2 obtains initial credit score value, with
Judge whether to borrow or lend money, such as specifically using 0.5 as discrimination threshold, when initial credit score value p >=0.5, is then determined as borrower
It can borrow or lend money, initial credit score value p<When 0.5, then it is determined as that borrower can not provide a loan.
In the present embodiment, the correspondence between credit scoring value and credit line is specially：
Wherein, P is initial credit score value, and Q is line of credit angle value and f (Q) is probability-distribution function, as shown in Figure 2.
After initial credit score value is calculated in step S2, corresponding credit line Q can be calculated by formula (4)
Value.
Further include the correspondence pre-established between credit scoring value and line of credit angle value in the present embodiment, in step S3
Table calculates to after line of credit angle value, corresponding credit scoring value is found out from mapping table, it is final to obtain user to be assessed
Credit scoring value output.
Above-mentioned only presently preferred embodiments of the present invention, is not intended to limit the present invention in any form.Although of the invention
Disclosed above with preferred embodiment, however, it is not intended to limit the invention.Therefore, every without departing from technical solution of the present invention
Content, technical spirit any simple modifications, equivalents, and modifications made to the above embodiment, should all fall according to the present invention
In the range of technical solution of the present invention protection.
Claims (9)
1. a kind of credit line intelligent evaluation method, which is characterized in that step includes：
S1., user is normally fulfiled to the probability of happening p of agreement to train as initial credit score value, and using user credit data
The model of Probability p obtains initial credit Rating Model；
S2. the credit data for obtaining user to be assessed carries out the credit data of acquisition using the initial credit Rating Model
Scoring, obtains initial credit score value；
S3., the initial credit score value is input to the credit scoring value and line of credit built in advance according to probability density characteristics
In probability distribution relationship between angle value, the line of credit angle value of user to be assessed is calculated.
2. credit line intelligent evaluation method according to claim 1, it is characterised in that：It is specifically based in the step S1
Logic Regression Models train the model of Probability p, obtain initial credit Rating Model.
3. credit line intelligent evaluation method according to claim 2, it is characterised in that：The specific steps of the step S1
For：
S11. it obtains original user credit data and carries out data prediction, data after output pretreatment；
S12. discretization data in data after the pretreatment are subjected to numeralization processing, to use numerical identity items discretization
The attribute of data, obtains numeric type data；
S13. logic-based regression model is trained the obtained numeric type data, obtains logic-based regression model
Initial credit Rating Model.
4. credit line intelligent evaluation method according to claim 3, it is characterised in that：Into line number in the step S11
It when Data preprocess, specifically searches missing values, redundancy value and exceptional value in the original user credit data, and will find
The missing values carry out Lagrange's interpolation processing, and the redundancy value found, exceptional value are carried out delete processing.
5. credit line intelligent evaluation method according to claim 4, it is characterised in that：The step S12 will be pre-processed
When discretization data carry out numeralization processing in data afterwards, the data with incremental relationship type are specifically used into incremental numerical value mark
Know, adds multiple attributes respectively without the data for being incremented by relationship type, each attribute is identified using numerical value.
6. the credit line intelligent evaluation method according to any one of claim 2~5, which is characterized in that the base
When Logic Regression Models build the credit scoring model, a logical transition first specifically is done to p：
L=β0+β1x1+βixi++βnxn=βTx
Wherein, i represents each attribute, and β is the weight coefficient of each attribute, xiFor the value under the attribute, β=(β0,β1,,βi,βn
)T, x=(x0,x1,,xi,xn)T；
The function model that final structure obtains p is：
The function model of the p obtained by above-mentioned structure is compressed to predicted value is exported between [0,1].
7. the credit line intelligent evaluation method according to any one of Claims 1 to 5, which is characterized in that the step
Rapid S2 the specific steps are：
S21. the original credit data for obtaining user to be assessed carries out data prediction, data after output pretreatment；
S22. data after pretreatment that the step S21 is obtained are subjected to numeralization processing, to use numerical identity items discrete
The attribute for changing data, obtains numeric type data；
S23. the obtained numeric type datas of the step S22 are input in the initial credit Rating Model, are initially believed
Use score value.
8. the credit line intelligent evaluation method according to any one of Claims 1 to 5, which is characterized in that the letter
It is normal distribution or Student distributions with the probability distribution relationship between score value and line of credit angle value, and meets：
Wherein, P is credit scoring value, and Q is line of credit angle value, and f (Q) is probability-distribution function.
9. credit line intelligent evaluation method according to claim 8, which is characterized in that further include pre- in the step S3
The mapping table first to build one's credit between score value and line of credit angle value is calculated to after the line of credit angle value, from described right
It answers and finds out corresponding credit scoring value in relation table, obtain the final credit scoring value output of user to be assessed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810415598.9A CN108615191A (en) | 2018-05-03 | 2018-05-03 | A kind of credit line intelligent evaluation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810415598.9A CN108615191A (en) | 2018-05-03 | 2018-05-03 | A kind of credit line intelligent evaluation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108615191A true CN108615191A (en) | 2018-10-02 |
Family
ID=63661815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810415598.9A Pending CN108615191A (en) | 2018-05-03 | 2018-05-03 | A kind of credit line intelligent evaluation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108615191A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767312A (en) * | 2018-12-10 | 2019-05-17 | 江西师范大学 | A kind of training of credit evaluation model, appraisal procedure and device |
| CN110060144A (en) * | 2019-03-18 | 2019-07-26 | 平安科技（深圳）有限公司 | Amount model training method, amount appraisal procedure, device, equipment and medium |
| CN110175910A (en) * | 2019-05-31 | 2019-08-27 | 阿里巴巴集团控股有限公司 | Handle the method, apparatus and electronic equipment of credit service request |
| CN111598677A (en) * | 2020-07-24 | 2020-08-28 | 北京淇瑀信息科技有限公司 | Resource quota determining method and device and electronic equipment |
| CN111899088A (en) * | 2020-06-23 | 2020-11-06 | 四川新网银行股份有限公司 | Accurate asset limit calculation method under high-concurrency data flow field scene |
| CN113177701A (en) * | 2021-04-15 | 2021-07-27 | 国任财产保险股份有限公司 | User credit assessment method and device |
| CN113487410A (en) * | 2021-07-06 | 2021-10-08 | 建信金融科技有限责任公司 | Credit granting management method and device, electronic equipment and computer readable medium |
-
2018
- 2018-05-03 CN CN201810415598.9A patent/CN108615191A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767312A (en) * | 2018-12-10 | 2019-05-17 | 江西师范大学 | A kind of training of credit evaluation model, appraisal procedure and device |
| CN109767312B (en) * | 2018-12-10 | 2023-05-09 | 江西师范大学 | Credit evaluation model training and evaluation method and device |
| CN110060144A (en) * | 2019-03-18 | 2019-07-26 | 平安科技（深圳）有限公司 | Amount model training method, amount appraisal procedure, device, equipment and medium |
| CN110060144B (en) * | 2019-03-18 | 2024-01-30 | 平安科技（深圳）有限公司 | Method for training credit model, method, device, equipment and medium for evaluating credit |
| CN110175910A (en) * | 2019-05-31 | 2019-08-27 | 阿里巴巴集团控股有限公司 | Handle the method, apparatus and electronic equipment of credit service request |
| CN111899088A (en) * | 2020-06-23 | 2020-11-06 | 四川新网银行股份有限公司 | Accurate asset limit calculation method under high-concurrency data flow field scene |
| CN111899088B (en) * | 2020-06-23 | 2023-04-18 | 四川新网银行股份有限公司 | Accurate asset limit calculation method under high-concurrency data flow field scene |
| CN111598677A (en) * | 2020-07-24 | 2020-08-28 | 北京淇瑀信息科技有限公司 | Resource quota determining method and device and electronic equipment |
| CN113177701A (en) * | 2021-04-15 | 2021-07-27 | 国任财产保险股份有限公司 | User credit assessment method and device |
| CN113487410A (en) * | 2021-07-06 | 2021-10-08 | 建信金融科技有限责任公司 | Credit granting management method and device, electronic equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108615191A (en) | A kind of credit line intelligent evaluation method | |
| Jin et al. | A data-driven approach to predict default risk of loan for online peer-to-peer (P2P) lending | |
| Schneider et al. | Infer, predict, and assure: Accounting opportunities in data analytics | |
| Tang et al. | The determinants of ESG ratings: Rater ownership matters | |
| CN107424070A (en) | A kind of loan user credit ranking method and system based on machine learning | |
| Ereiz | Predicting default loans using machine learning (OptiML) | |
| CN108492001A (en) | A method of being used for guaranteed loan network risk management | |
| Purwidianti | An empirical study on family financial behavior | |
| Atkeson et al. | Industry evolution and transition: Measuring investment in organization capital | |
| Mousseau et al. | On the notion of category size in multiple criteria sorting models | |
| CN114862563A (en) | Small and medium credit strategy model based on principal component analysis and neural network | |
| CN110675240B (en) | Monitoring method and system for risk radar early warning | |
| Panwai | Artificial neural network stock price prediction model under the influence of big data | |
| Bozsik et al. | Decision tree-based credit decision support system | |
| CN111694952A (en) | Big data analysis model system based on microblog and implementation method thereof | |
| KR102576143B1 (en) | Method for performing continual learning on credit scoring without reject inference and recording medium recording computer readable program for executing the method | |
| Zagurskiy et al. | Management Models and Evaluation of Reputation Risks | |
| Penalver | Capital flows to emerging markets | |
| Soares et al. | A simple fuzzy system applied to predict default rate | |
| Weitzman et al. | ITGR Integer Holdings Corporation Common Stock | |
| Breeden | Solving the Long-Range Forecasting Problem in Supervised Learning | |
| Jorgenson et al. | ORI Old Republic International Corporation Common Stock | |
| TWM641546U (en) | Credit risk stress test and management system | |
| Weitzman et al. | Is BLEU Stock Expected to Go Up? | |
| ZXhang et al. | HCP HashiCorp Inc. Class A Common Stock |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200514 Address after: Guanxi Town, Dingcheng District, Changde, Hunan Province Applicant after: Hunan Huda Jinke Technology Development Co.,Ltd. Address before: Yuelu District City, Hunan province 410082 Changsha Lushan South Road, Hunan University College of information science and Engineering Applicant before: HUNAN University |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181002 |
|
| RJ01 | Rejection of invention patent application after publication |